id	sid	tid	token	lemma	pos
cana-5407	1	1	communications	communication	NOUN
cana-5407	1	2	on	on	ADP
cana-5407	1	3	applied	apply	VERB
cana-5407	1	4	nonlinear	nonlinear	ADJ
cana-5407	1	5	analysis	analysis	NOUN
cana-5407	1	6	issn	issn	NOUN
cana-5407	1	7	:	:	PUNCT
cana-5407	1	8	1074	1074	NUM
cana-5407	1	9	-	-	PUNCT
cana-5407	1	10	133x	133x	NUM
cana-5407	1	11	vol	vol	VERB
cana-5407	1	12	32	32	NUM
cana-5407	1	13	no	no	NOUN
cana-5407	1	14	.	.	PUNCT
cana-5407	2	1	10s	10	NOUN
cana-5407	2	2	(	(	PUNCT
cana-5407	2	3	2025	2025	NUM
cana-5407	2	4	)	)	PUNCT
cana-5407	2	5	2172	2172	NUM
cana-5407	2	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-5407	2	7	voice	voice	NOUN
cana-5407	2	8	conversion	conversion	NOUN
cana-5407	2	9	using	use	VERB
cana-5407	2	10	hybrid	hybrid	ADJ
cana-5407	2	11	cnn	cnn	PROPN
cana-5407	2	12	bilstm	bilstm	PROPN
cana-5407	2	13	-	-	PUNCT
cana-5407	2	14	wavenet	wavenet	NOUN
cana-5407	2	15	deep	deep	ADJ
cana-5407	2	16	learning	learning	NOUN
cana-5407	2	17	models	model	NOUN
cana-5407	2	18	a.	a.	PROPN
cana-5407	2	19	bala	bala	PROPN
cana-5407	2	20	raju1	raju1	PROPN
cana-5407	2	21	,	,	PUNCT
cana-5407	2	22	s.	s.	PROPN
cana-5407	2	23	p	p	PROPN
cana-5407	2	24	singh2	singh2	PROPN
cana-5407	2	25	,	,	PUNCT
cana-5407	2	26	dhiraj	dhiraj	ADJ
cana-5407	2	27	sunehra3	sunehra3	PROPN
cana-5407	2	28	1research	1research	NUM
cana-5407	2	29	scholar	scholar	NOUN
cana-5407	2	30	,	,	PUNCT
cana-5407	2	31	department	department	NOUN
cana-5407	2	32	of	of	ADP
cana-5407	2	33	electronics	electronic	NOUN
cana-5407	2	34	and	and	CCONJ
cana-5407	2	35	communication	communication	NOUN
cana-5407	2	36	engineering	engineering	NOUN
cana-5407	2	37	,	,	PUNCT
cana-5407	2	38	jntuh	jntuh	PROPN
cana-5407	2	39	,	,	PUNCT
cana-5407	2	40	kukatpally	kukatpally	ADV
cana-5407	2	41	,	,	PUNCT
cana-5407	2	42	hyderabad	hyderabad	PROPN
cana-5407	2	43	.	.	PUNCT
cana-5407	3	1	1assistant	1assistant	NUM
cana-5407	3	2	professor	professor	NOUN
cana-5407	3	3	,	,	PUNCT
cana-5407	3	4	department	department	NOUN
cana-5407	3	5	of	of	ADP
cana-5407	3	6	electronics	electronic	NOUN
cana-5407	3	7	and	and	CCONJ
cana-5407	3	8	communication	communication	NOUN
cana-5407	3	9	engineering	engineering	NOUN
cana-5407	3	10	,	,	PUNCT
cana-5407	3	11	mahatma	mahatma	PROPN
cana-5407	3	12	gandhi	gandhi	PROPN
cana-5407	3	13	institute	institute	PROPN
cana-5407	3	14	of	of	ADP
cana-5407	3	15	technology	technology	PROPN
cana-5407	3	16	,	,	PUNCT
cana-5407	3	17	hyderabad	hyderabad	PROPN
cana-5407	3	18	,	,	PUNCT
cana-5407	3	19	telangana	telangana	PROPN
cana-5407	3	20	,	,	PUNCT
cana-5407	3	21	india	india	PROPN
cana-5407	3	22	.	.	PUNCT
cana-5407	4	1	2professor	2professor	NUM
cana-5407	4	2	,	,	PUNCT
cana-5407	4	3	department	department	NOUN
cana-5407	4	4	of	of	ADP
cana-5407	4	5	electronics	electronic	NOUN
cana-5407	4	6	and	and	CCONJ
cana-5407	4	7	communication	communication	NOUN
cana-5407	4	8	engineering	engineering	NOUN
cana-5407	4	9	,	,	PUNCT
cana-5407	4	10	mahatma	mahatma	PROPN
cana-5407	4	11	gandhi	gandhi	PROPN
cana-5407	4	12	institute	institute	PROPN
cana-5407	4	13	of	of	ADP
cana-5407	4	14	technology	technology	PROPN
cana-5407	4	15	,	,	PUNCT
cana-5407	4	16	hyderabad	hyderabad	PROPN
cana-5407	4	17	,	,	PUNCT
cana-5407	4	18	telangana	telangana	PROPN
cana-5407	4	19	,	,	PUNCT
cana-5407	4	20	india	india	PROPN
cana-5407	4	21	.	.	PUNCT
cana-5407	5	1	3professor	3professor	NUM
cana-5407	5	2	,	,	PUNCT
cana-5407	5	3	department	department	NOUN
cana-5407	5	4	of	of	ADP
cana-5407	5	5	electronics	electronic	NOUN
cana-5407	5	6	and	and	CCONJ
cana-5407	5	7	communication	communication	NOUN
cana-5407	5	8	engineering	engineering	NOUN
cana-5407	5	9	,	,	PUNCT
cana-5407	5	10	jntuh	jntuh	NOUN
cana-5407	5	11	ucer	ucer	NOUN
cana-5407	5	12	,	,	PUNCT
cana-5407	5	13	rajanna	rajanna	NOUN
cana-5407	5	14	sircilla	sircilla	NOUN
cana-5407	5	15	,	,	PUNCT
cana-5407	5	16	telangana	telangana	PROPN
cana-5407	5	17	,	,	PUNCT
cana-5407	5	18	india	india	PROPN
cana-5407	5	19	.	.	PUNCT
cana-5407	5	20	balaraju_ap@yahoo.co.in1	balaraju_ap@yahoo.co.in1	PROPN
cana-5407	5	21	,	,	PUNCT
cana-5407	5	22	spsingh@mgit.ac.in2	spsingh@mgit.ac.in2	PROPN
cana-5407	5	23	,	,	PUNCT
cana-5407	5	24	dhirajsunehra@yahoo.co.in3	dhirajsunehra@yahoo.co.in3	X
cana-5407	5	25	article	article	NOUN
cana-5407	5	26	history	history	NOUN
cana-5407	5	27	:	:	PUNCT
cana-5407	5	28	received	receive	VERB
cana-5407	5	29	:	:	PUNCT
cana-5407	5	30	12	12	NUM
cana-5407	5	31	-	-	SYM
cana-5407	5	32	01	01	NUM
cana-5407	5	33	-	-	PUNCT
cana-5407	5	34	2025	2025	NUM
cana-5407	5	35	revised	revise	VERB
cana-5407	5	36	:	:	PUNCT
cana-5407	5	37	15	15	NUM
cana-5407	5	38	-	-	NUM
cana-5407	5	39	02	02	NUM
cana-5407	5	40	-	-	PUNCT
cana-5407	5	41	2025	2025	NUM
cana-5407	5	42	accepted	accept	VERB
cana-5407	5	43	:	:	PUNCT
cana-5407	5	44	01	01	NUM
cana-5407	5	45	-	-	SYM
cana-5407	5	46	03	03	NUM
cana-5407	5	47	-	-	PUNCT
cana-5407	5	48	2025	2025	NUM
cana-5407	5	49	abstract	abstract	NOUN
cana-5407	5	50	:	:	PUNCT
cana-5407	5	51	voice	voice	NOUN
cana-5407	5	52	conversion	conversion	NOUN
cana-5407	5	53	is	be	AUX
cana-5407	5	54	an	an	DET
cana-5407	5	55	exciting	exciting	ADJ
cana-5407	5	56	area	area	NOUN
cana-5407	5	57	of	of	ADP
cana-5407	5	58	speech	speech	NOUN
cana-5407	5	59	processing	processing	NOUN
cana-5407	5	60	in	in	ADP
cana-5407	5	61	which	which	PRON
cana-5407	5	62	deep	deep	ADJ
cana-5407	5	63	learning	learning	NOUN
cana-5407	5	64	approaches	approach	NOUN
cana-5407	5	65	are	be	AUX
cana-5407	5	66	developed	develop	VERB
cana-5407	5	67	that	that	PRON
cana-5407	5	68	can	can	AUX
cana-5407	5	69	modify	modify	VERB
cana-5407	5	70	the	the	DET
cana-5407	5	71	vocal	vocal	ADJ
cana-5407	5	72	qualities	quality	NOUN
cana-5407	5	73	of	of	ADP
cana-5407	5	74	an	an	DET
cana-5407	5	75	speaker	speaker	NOUN
cana-5407	5	76	to	to	PART
cana-5407	5	77	resemble	resemble	VERB
cana-5407	5	78	the	the	DET
cana-5407	5	79	voice	voice	NOUN
cana-5407	5	80	of	of	ADP
cana-5407	5	81	another	another	DET
cana-5407	5	82	person	person	NOUN
cana-5407	5	83	without	without	ADP
cana-5407	5	84	altering	alter	VERB
cana-5407	5	85	the	the	DET
cana-5407	5	86	context	context	NOUN
cana-5407	5	87	of	of	ADP
cana-5407	5	88	the	the	DET
cana-5407	5	89	utterance	utterance	NOUN
cana-5407	5	90	.	.	PUNCT
cana-5407	6	1	the	the	DET
cana-5407	6	2	significance	significance	NOUN
cana-5407	6	3	of	of	ADP
cana-5407	6	4	speech	speech	NOUN
cana-5407	6	5	conversion	conversion	NOUN
cana-5407	6	6	can	can	AUX
cana-5407	6	7	not	not	PART
cana-5407	6	8	be	be	AUX
cana-5407	6	9	overstated	overstate	VERB
cana-5407	6	10	,	,	PUNCT
cana-5407	6	11	as	as	SCONJ
cana-5407	6	12	it	it	PRON
cana-5407	6	13	is	be	AUX
cana-5407	6	14	employed	employ	VERB
cana-5407	6	15	in	in	ADP
cana-5407	6	16	a	a	DET
cana-5407	6	17	wide	wide	ADJ
cana-5407	6	18	range	range	NOUN
cana-5407	6	19	of	of	ADP
cana-5407	6	20	systems	system	NOUN
cana-5407	6	21	,	,	PUNCT
cana-5407	6	22	including	include	VERB
cana-5407	6	23	entertainment	entertainment	NOUN
cana-5407	6	24	,	,	PUNCT
cana-5407	6	25	vocal	vocal	ADJ
cana-5407	6	26	communication	communication	NOUN
cana-5407	6	27	,	,	PUNCT
cana-5407	6	28	and	and	CCONJ
cana-5407	6	29	privacy	privacy	NOUN
cana-5407	6	30	enhancement	enhancement	NOUN
cana-5407	6	31	.	.	PUNCT
cana-5407	7	1	however	however	ADV
cana-5407	7	2	,	,	PUNCT
cana-5407	7	3	traditional	traditional	ADJ
cana-5407	7	4	methods	method	NOUN
cana-5407	7	5	have	have	AUX
cana-5407	7	6	fallen	fall	VERB
cana-5407	7	7	short	short	ADJ
cana-5407	7	8	in	in	ADP
cana-5407	7	9	the	the	DET
cana-5407	7	10	face	face	NOUN
cana-5407	7	11	of	of	ADP
cana-5407	7	12	large	large	ADJ
cana-5407	7	13	data	datum	NOUN
cana-5407	7	14	sets	set	NOUN
cana-5407	7	15	and	and	CCONJ
cana-5407	7	16	the	the	DET
cana-5407	7	17	preservation	preservation	NOUN
cana-5407	7	18	of	of	ADP
cana-5407	7	19	subtle	subtle	ADJ
cana-5407	7	20	emotions	emotion	NOUN
cana-5407	7	21	,	,	PUNCT
cana-5407	7	22	hindering	hinder	VERB
cana-5407	7	23	voice	voice	NOUN
cana-5407	7	24	simulation	simulation	NOUN
cana-5407	7	25	.	.	PUNCT
cana-5407	8	1	to	to	PART
cana-5407	8	2	address	address	VERB
cana-5407	8	3	the	the	DET
cana-5407	8	4	above	above	ADJ
cana-5407	8	5	limitations	limitation	NOUN
cana-5407	8	6	,	,	PUNCT
cana-5407	8	7	we	we	PRON
cana-5407	8	8	present	present	VERB
cana-5407	8	9	a	a	DET
cana-5407	8	10	novel	novel	ADJ
cana-5407	8	11	way	way	NOUN
cana-5407	8	12	that	that	PRON
cana-5407	8	13	combines	combine	VERB
cana-5407	8	14	the	the	DET
cana-5407	8	15	fusion	fusion	NOUN
cana-5407	8	16	of	of	ADP
cana-5407	8	17	the	the	DET
cana-5407	8	18	speech	speech	NOUN
cana-5407	8	19	to	to	PART
cana-5407	8	20	text	text	VERB
cana-5407	8	21	technology	technology	NOUN
cana-5407	8	22	with	with	ADP
cana-5407	8	23	a	a	DET
cana-5407	8	24	text	text	NOUN
cana-5407	8	25	-	-	PUNCT
cana-5407	8	26	to	to	ADP
cana-5407	8	27	-	-	PUNCT
cana-5407	8	28	speech	speech	NOUN
cana-5407	8	29	transformation	transformation	NOUN
cana-5407	8	30	system	system	NOUN
cana-5407	8	31	powered	power	VERB
cana-5407	8	32	by	by	ADP
cana-5407	8	33	a	a	DET
cana-5407	8	34	deep	deep	ADJ
cana-5407	8	35	learning	learning	NOUN
cana-5407	8	36	architecture	architecture	NOUN
cana-5407	8	37	.	.	PUNCT
cana-5407	9	1	the	the	DET
cana-5407	9	2	system	system	NOUN
cana-5407	9	3	contains	contain	VERB
cana-5407	9	4	advanced	advanced	ADJ
cana-5407	9	5	embedding	embed	VERB
cana-5407	9	6	layers	layer	NOUN
cana-5407	9	7	like	like	ADP
cana-5407	9	8	phoneme	phoneme	NOUN
cana-5407	9	9	embedding	embed	VERB
cana-5407	9	10	,	,	PUNCT
cana-5407	9	11	bidirectional	bidirectional	ADJ
cana-5407	9	12	long	long	ADJ
cana-5407	9	13	short	short	ADJ
cana-5407	9	14	-	-	PUNCT
cana-5407	9	15	term	term	NOUN
cana-5407	9	16	memory	memory	NOUN
cana-5407	9	17	(	(	PUNCT
cana-5407	9	18	lstm	lstm	NOUN
cana-5407	9	19	)	)	PUNCT
cana-5407	9	20	networks	network	NOUN
cana-5407	9	21	,	,	PUNCT
cana-5407	9	22	and	and	CCONJ
cana-5407	9	23	wavenet	wavenet	ADJ
cana-5407	9	24	vocoder	vocoder	NOUN
cana-5407	9	25	,	,	PUNCT
cana-5407	9	26	which	which	PRON
cana-5407	9	27	make	make	VERB
cana-5407	9	28	the	the	DET
cana-5407	9	29	transformed	transform	VERB
cana-5407	9	30	voice	voice	NOUN
cana-5407	9	31	more	more	ADV
cana-5407	9	32	accurate	accurate	ADJ
cana-5407	9	33	and	and	CCONJ
cana-5407	9	34	authentic	authentic	ADJ
cana-5407	9	35	.	.	PUNCT
cana-5407	10	1	in	in	ADP
cana-5407	10	2	the	the	DET
cana-5407	10	3	proposed	propose	VERB
cana-5407	10	4	model	model	NOUN
cana-5407	10	5	,	,	PUNCT
cana-5407	10	6	we	we	PRON
cana-5407	10	7	use	use	VERB
cana-5407	10	8	the	the	DET
cana-5407	10	9	speech	speech	NOUN
cana-5407	10	10	recognition	recognition	NOUN
cana-5407	10	11	tools	tool	NOUN
cana-5407	10	12	packages	package	NOUN
cana-5407	10	13	of	of	ADP
cana-5407	10	14	python	python	NOUN
cana-5407	10	15	and	and	CCONJ
cana-5407	10	16	complex	complex	ADJ
cana-5407	10	17	neural	neural	ADJ
cana-5407	10	18	network	network	NOUN
cana-5407	10	19	methods	method	NOUN
cana-5407	10	20	to	to	PART
cana-5407	10	21	improve	improve	VERB
cana-5407	10	22	the	the	DET
cana-5407	10	23	naturalness	naturalness	NOUN
cana-5407	10	24	and	and	CCONJ
cana-5407	10	25	clarity	clarity	NOUN
cana-5407	10	26	.	.	PUNCT
cana-5407	11	1	moreover	moreover	ADV
cana-5407	11	2	,	,	PUNCT
cana-5407	11	3	it	it	PRON
cana-5407	11	4	sets	set	VERB
cana-5407	11	5	a	a	DET
cana-5407	11	6	bar	bar	NOUN
cana-5407	11	7	when	when	SCONJ
cana-5407	11	8	it	it	PRON
cana-5407	11	9	comes	come	VERB
cana-5407	11	10	to	to	ADP
cana-5407	11	11	processing	processing	NOUN
cana-5407	11	12	power	power	NOUN
cana-5407	11	13	,	,	PUNCT
cana-5407	11	14	efficiency	efficiency	NOUN
cana-5407	11	15	,	,	PUNCT
cana-5407	11	16	and	and	CCONJ
cana-5407	11	17	performance	performance	NOUN
cana-5407	11	18	.	.	PUNCT
cana-5407	12	1	keywords	keyword	NOUN
cana-5407	12	2	:	:	PUNCT
cana-5407	12	3	voice	voice	NOUN
cana-5407	12	4	conversion	conversion	NOUN
cana-5407	12	5	,	,	PUNCT
cana-5407	12	6	deep	deep	ADJ
cana-5407	12	7	learning	learning	NOUN
cana-5407	12	8	,	,	PUNCT
cana-5407	12	9	speech	speech	NOUN
cana-5407	12	10	processing	processing	NOUN
cana-5407	12	11	,	,	PUNCT
cana-5407	12	12	speech	speech	NOUN
cana-5407	12	13	to	to	ADP
cana-5407	12	14	text	text	NOUN
cana-5407	12	15	,	,	PUNCT
cana-5407	12	16	bidirectional	bidirectional	ADJ
cana-5407	12	17	lstm	lstm	PROPN
cana-5407	12	18	,	,	PUNCT
cana-5407	12	19	wavenet	wavenet	ADJ
cana-5407	12	20	vocoder	vocoder	NOUN
cana-5407	12	21	introduction	introduction	NOUN
cana-5407	12	22	:	:	PUNCT
cana-5407	12	23	voice	voice	NOUN
cana-5407	12	24	conversion	conversion	NOUN
cana-5407	12	25	is	be	AUX
cana-5407	12	26	a	a	DET
cana-5407	12	27	domain	domain	NOUN
cana-5407	12	28	of	of	ADP
cana-5407	12	29	speech	speech	NOUN
cana-5407	12	30	processing	processing	NOUN
cana-5407	12	31	,	,	PUNCT
cana-5407	12	32	and	and	CCONJ
cana-5407	12	33	it	it	PRON
cana-5407	12	34	is	be	AUX
cana-5407	12	35	fascinating	fascinating	ADJ
cana-5407	12	36	to	to	PART
cana-5407	12	37	interact	interact	VERB
cana-5407	12	38	by	by	ADP
cana-5407	12	39	observing	observe	VERB
cana-5407	12	40	how	how	SCONJ
cana-5407	12	41	the	the	DET
cana-5407	12	42	special	special	ADJ
cana-5407	12	43	human	human	ADJ
cana-5407	12	44	speech	speech	NOUN
cana-5407	12	45	ability	ability	NOUN
cana-5407	12	46	specifies	specifie	NOUN
cana-5407	12	47	in	in	ADP
cana-5407	12	48	words	word	NOUN
cana-5407	12	49	is	be	AUX
cana-5407	12	50	first	first	ADV
cana-5407	12	51	converted	convert	VERB
cana-5407	12	52	into	into	ADP
cana-5407	12	53	a	a	DET
cana-5407	12	54	different	different	ADJ
cana-5407	12	55	form	form	NOUN
cana-5407	12	56	that	that	PRON
cana-5407	12	57	maintains	maintain	VERB
cana-5407	12	58	the	the	DET
cana-5407	12	59	semantic	semantic	ADJ
cana-5407	12	60	bearing	bearing	NOUN
cana-5407	12	61	of	of	ADP
cana-5407	12	62	what	what	PRON
cana-5407	12	63	the	the	DET
cana-5407	12	64	original	original	ADJ
cana-5407	12	65	voice	voice	NOUN
cana-5407	12	66	has	have	AUX
cana-5407	12	67	spoken	speak	VERB
cana-5407	12	68	[	[	PUNCT
cana-5407	12	69	1	1	NUM
cana-5407	12	70	]	]	PUNCT
cana-5407	12	71	.	.	PUNCT
cana-5407	13	1	this	this	PRON
cana-5407	13	2	means	mean	VERB
cana-5407	13	3	that	that	SCONJ
cana-5407	13	4	the	the	DET
cana-5407	13	5	voice	voice	NOUN
cana-5407	13	6	of	of	ADP
cana-5407	13	7	the	the	DET
cana-5407	13	8	person	person	NOUN
cana-5407	13	9	is	be	AUX
cana-5407	13	10	converted	convert	VERB
cana-5407	13	11	back	back	ADV
cana-5407	13	12	to	to	ADP
cana-5407	13	13	another	another	DET
cana-5407	13	14	person	person	NOUN
cana-5407	13	15	’s	’s	PART
cana-5407	13	16	voice	voice	NOUN
cana-5407	13	17	with	with	ADP
cana-5407	13	18	different	different	ADJ
cana-5407	13	19	pitch	pitch	NOUN
cana-5407	13	20	,	,	PUNCT
cana-5407	13	21	voice	voice	NOUN
cana-5407	13	22	,	,	PUNCT
cana-5407	13	23	and	and	CCONJ
cana-5407	13	24	accent	accent	ADJ
cana-5407	13	25	.	.	PUNCT
cana-5407	14	1	this	this	DET
cana-5407	14	2	technology	technology	NOUN
cana-5407	14	3	has	have	AUX
cana-5407	14	4	employed	employ	VERB
cana-5407	14	5	a	a	DET
cana-5407	14	6	combination	combination	NOUN
cana-5407	14	7	of	of	ADP
cana-5407	14	8	methods	method	NOUN
cana-5407	14	9	,	,	PUNCT
cana-5407	14	10	including	include	VERB
cana-5407	14	11	digital	digital	ADJ
cana-5407	14	12	signal	signal	NOUN
cana-5407	14	13	processing	processing	NOUN
cana-5407	14	14	.	.	PUNCT
cana-5407	15	1	in	in	ADP
cana-5407	15	2	recent	recent	ADJ
cana-5407	15	3	years	year	NOUN
cana-5407	15	4	,	,	PUNCT
cana-5407	15	5	its	its	PRON
cana-5407	15	6	accuracy	accuracy	NOUN
cana-5407	15	7	has	have	AUX
cana-5407	15	8	set	set	VERB
cana-5407	15	9	new	new	ADJ
cana-5407	15	10	records	record	NOUN
cana-5407	15	11	,	,	PUNCT
cana-5407	15	12	mainly	mainly	ADV
cana-5407	15	13	due	due	ADP
cana-5407	15	14	to	to	ADP
cana-5407	15	15	new	new	ADJ
cana-5407	15	16	deep	deep	ADJ
cana-5407	15	17	learning	learning	NOUN
cana-5407	15	18	techniques	technique	NOUN
cana-5407	15	19	[	[	X
cana-5407	15	20	2	2	NUM
cana-5407	15	21	]	]	PUNCT
cana-5407	15	22	.	.	PUNCT
cana-5407	16	1	voice	voice	NOUN
cana-5407	16	2	conversion	conversion	NOUN
cana-5407	16	3	has	have	AUX
cana-5407	16	4	been	be	AUX
cana-5407	16	5	vital	vital	ADJ
cana-5407	16	6	in	in	ADP
cana-5407	16	7	several	several	ADJ
cana-5407	16	8	applications	application	NOUN
cana-5407	16	9	,	,	PUNCT
cana-5407	16	10	such	such	ADJ
cana-5407	16	11	as	as	ADP
cana-5407	16	12	:	:	PUNCT
cana-5407	16	13	entertainment	entertainment	NOUN
cana-5407	16	14	–	–	PUNCT
cana-5407	16	15	it	it	PRON
cana-5407	16	16	has	have	AUX
cana-5407	16	17	used	use	VERB
cana-5407	16	18	the	the	DET
cana-5407	16	19	technology	technology	NOUN
cana-5407	16	20	to	to	PART
cana-5407	16	21	overwrite	overwrite	VERB
cana-5407	16	22	films	film	NOUN
cana-5407	16	23	and	and	CCONJ
cana-5407	16	24	video	video	NOUN
cana-5407	16	25	games	game	NOUN
cana-5407	16	26	when	when	SCONJ
cana-5407	16	27	actors	actor	NOUN
cana-5407	16	28	’	’	PART
cana-5407	16	29	voices	voice	NOUN
cana-5407	16	30	need	need	VERB
cana-5407	16	31	changing	change	VERB
cana-5407	16	32	and	and	CCONJ
cana-5407	16	33	produce	produce	VERB
cana-5407	16	34	multi	multi	ADJ
cana-5407	16	35	-	-	ADJ
cana-5407	16	36	voiced	voice	VERB
cana-5407	16	37	characters	character	NOUN
cana-5407	16	38	[	[	X
cana-5407	16	39	3	3	NUM
cana-5407	16	40	]	]	PUNCT
cana-5407	16	41	.	.	PUNCT
cana-5407	17	1	another	another	DET
cana-5407	17	2	application	application	NOUN
cana-5407	17	3	is	be	AUX
cana-5407	17	4	in	in	ADP
cana-5407	17	5	communication	communication	NOUN
cana-5407	17	6	,	,	PUNCT
cana-5407	17	7	where	where	SCONJ
cana-5407	17	8	the	the	DET
cana-5407	17	9	speech	speech	NOUN
cana-5407	17	10	becomes	become	VERB
cana-5407	17	11	communications	communication	NOUN
cana-5407	17	12	on	on	ADP
cana-5407	17	13	applied	apply	VERB
cana-5407	17	14	nonlinear	nonlinear	ADJ
cana-5407	17	15	analysis	analysis	NOUN
cana-5407	17	16	issn	issn	NOUN
cana-5407	17	17	:	:	PUNCT
cana-5407	17	18	1074	1074	NUM
cana-5407	17	19	-	-	PUNCT
cana-5407	17	20	133x	133x	NUM
cana-5407	17	21	vol	vol	VERB
cana-5407	17	22	32	32	NUM
cana-5407	17	23	no	no	NOUN
cana-5407	17	24	.	.	PUNCT
cana-5407	18	1	10s	10	NOUN
cana-5407	18	2	(	(	PUNCT
cana-5407	18	3	2025	2025	NUM
cana-5407	18	4	)	)	PUNCT
cana-5407	18	5	2173	2173	NUM
cana-5407	18	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-5407	18	7	clearer	clear	ADJ
cana-5407	18	8	;	;	PUNCT
cana-5407	18	9	people	people	NOUN
cana-5407	18	10	’s	’s	PART
cana-5407	18	11	voices	voice	NOUN
cana-5407	18	12	are	be	AUX
cana-5407	18	13	better	well	ADV
cana-5407	18	14	understood	understand	VERB
cana-5407	18	15	using	use	VERB
cana-5407	18	16	this	this	DET
cana-5407	18	17	technology	technology	NOUN
cana-5407	18	18	on	on	ADP
cana-5407	18	19	all	all	DET
cana-5407	18	20	channels	channel	NOUN
cana-5407	18	21	and	and	CCONJ
cana-5407	18	22	in	in	ADP
cana-5407	18	23	a	a	DET
cana-5407	18	24	variety	variety	NOUN
cana-5407	18	25	of	of	ADP
cana-5407	18	26	languages	language	NOUN
cana-5407	18	27	and	and	CCONJ
cana-5407	18	28	dialects	dialect	NOUN
cana-5407	18	29	.	.	PUNCT
cana-5407	19	1	another	another	DET
cana-5407	19	2	critical	critical	ADJ
cana-5407	19	3	application	application	NOUN
cana-5407	19	4	of	of	ADP
cana-5407	19	5	voice	voice	NOUN
cana-5407	19	6	conversion	conversion	NOUN
cana-5407	19	7	is	be	AUX
cana-5407	19	8	personal	personal	ADJ
cana-5407	19	9	safety	safety	NOUN
cana-5407	19	10	and	and	CCONJ
cana-5407	19	11	privacy	privacy	NOUN
cana-5407	19	12	[	[	X
cana-5407	19	13	4	4	NUM
cana-5407	19	14	]	]	PUNCT
cana-5407	19	15	.	.	PUNCT
cana-5407	20	1	since	since	SCONJ
cana-5407	20	2	it	it	PRON
cana-5407	20	3	does	do	AUX
cana-5407	20	4	not	not	PART
cana-5407	20	5	consider	consider	VERB
cana-5407	20	6	a	a	DET
cana-5407	20	7	person	person	NOUN
cana-5407	20	8	speaking	speak	VERB
cana-5407	20	9	a	a	DET
cana-5407	20	10	different	different	ADJ
cana-5407	20	11	language	language	NOUN
cana-5407	20	12	,	,	PUNCT
cana-5407	20	13	the	the	DET
cana-5407	20	14	problem	problem	NOUN
cana-5407	20	15	is	be	AUX
cana-5407	20	16	solved	solve	VERB
cana-5407	20	17	when	when	SCONJ
cana-5407	20	18	talking	talk	VERB
cana-5407	20	19	to	to	ADP
cana-5407	20	20	a	a	DET
cana-5407	20	21	person	person	NOUN
cana-5407	20	22	with	with	ADP
cana-5407	20	23	speech	speech	NOUN
cana-5407	20	24	difficulty	difficulty	NOUN
cana-5407	20	25	.	.	PUNCT
cana-5407	21	1	therefore	therefore	ADV
cana-5407	21	2	,	,	PUNCT
cana-5407	21	3	it	it	PRON
cana-5407	21	4	allows	allow	VERB
cana-5407	21	5	for	for	ADP
cana-5407	21	6	anonymization	anonymization	NOUN
cana-5407	21	7	in	in	ADP
cana-5407	21	8	sensitive	sensitive	ADJ
cana-5407	21	9	situations	situation	NOUN
cana-5407	21	10	.	.	PUNCT
cana-5407	22	1	currently	currently	ADV
cana-5407	22	2	,	,	PUNCT
cana-5407	22	3	most	most	ADJ
cana-5407	22	4	methods	method	NOUN
cana-5407	22	5	used	use	VERB
cana-5407	22	6	for	for	ADP
cana-5407	22	7	converting	convert	VERB
cana-5407	22	8	voices	voice	NOUN
cana-5407	22	9	are	be	AUX
cana-5407	22	10	based	base	VERB
cana-5407	22	11	on	on	ADP
cana-5407	22	12	deep	deep	ADJ
cana-5407	22	13	learning	learning	NOUN
cana-5407	22	14	models	model	NOUN
cana-5407	22	15	[	[	X
cana-5407	22	16	5	5	NUM
cana-5407	22	17	]	]	PUNCT
cana-5407	22	18	.	.	PUNCT
cana-5407	23	1	such	such	ADJ
cana-5407	23	2	models	model	NOUN
cana-5407	23	3	,	,	PUNCT
cana-5407	23	4	known	know	VERB
cana-5407	23	5	as	as	ADP
cana-5407	23	6	convolutional	convolutional	ADJ
cana-5407	23	7	neural	neural	ADJ
cana-5407	23	8	networks	network	NOUN
cana-5407	23	9	(	(	PUNCT
cana-5407	23	10	cnn	cnn	PROPN
cana-5407	23	11	)	)	PUNCT
cana-5407	23	12	,	,	PUNCT
cana-5407	23	13	recurrent	recurrent	ADJ
cana-5407	23	14	neural	neural	ADJ
cana-5407	23	15	networks	network	NOUN
cana-5407	23	16	(	(	PUNCT
cana-5407	23	17	rnn	rnn	PROPN
cana-5407	23	18	)	)	PUNCT
cana-5407	23	19	,	,	PUNCT
cana-5407	23	20	and	and	CCONJ
cana-5407	23	21	generative	generative	VERB
cana-5407	23	22	adversarial	adversarial	ADJ
cana-5407	23	23	networks	network	NOUN
cana-5407	23	24	(	(	PUNCT
cana-5407	23	25	gan	gan	PROPN
cana-5407	23	26	)	)	PUNCT
cana-5407	23	27	,	,	PUNCT
cana-5407	23	28	have	have	AUX
cana-5407	23	29	only	only	ADV
cana-5407	23	30	been	be	AUX
cana-5407	23	31	introduced	introduce	VERB
cana-5407	23	32	a	a	DET
cana-5407	23	33	decade	decade	NOUN
cana-5407	23	34	ago	ago	ADV
cana-5407	23	35	.	.	PUNCT
cana-5407	24	1	thanks	thank	NOUN
cana-5407	24	2	to	to	ADP
cana-5407	24	3	learning	learn	VERB
cana-5407	24	4	intricate	intricate	ADJ
cana-5407	24	5	patterns	pattern	NOUN
cana-5407	24	6	and	and	CCONJ
cana-5407	24	7	variations	variation	NOUN
cana-5407	24	8	from	from	ADP
cana-5407	24	9	huge	huge	ADJ
cana-5407	24	10	datasets	dataset	NOUN
cana-5407	24	11	of	of	ADP
cana-5407	24	12	speech	speech	NOUN
cana-5407	24	13	,	,	PUNCT
cana-5407	24	14	these	these	DET
cana-5407	24	15	technologies	technology	NOUN
cana-5407	24	16	have	have	AUX
cana-5407	24	17	radically	radically	ADV
cana-5407	24	18	improved	improve	VERB
cana-5407	24	19	the	the	DET
cana-5407	24	20	quality	quality	NOUN
cana-5407	24	21	and	and	CCONJ
cana-5407	24	22	the	the	DET
cana-5407	24	23	level	level	NOUN
cana-5407	24	24	of	of	ADP
cana-5407	24	25	voices	voice	NOUN
cana-5407	24	26	’	'	PUNCT
cana-5407	24	27	naturalness	naturalness	NOUN
cana-5407	24	28	that	that	PRON
cana-5407	24	29	are	be	AUX
cana-5407	24	30	converted	convert	VERB
cana-5407	24	31	[	[	X
cana-5407	24	32	6	6	NUM
cana-5407	24	33	]	]	PUNCT
cana-5407	24	34	.	.	PUNCT
cana-5407	25	1	while	while	SCONJ
cana-5407	25	2	a	a	DET
cana-5407	25	3	great	great	ADJ
cana-5407	25	4	deal	deal	NOUN
cana-5407	25	5	of	of	ADP
cana-5407	25	6	improvement	improvement	NOUN
cana-5407	25	7	has	have	AUX
cana-5407	25	8	been	be	AUX
cana-5407	25	9	achieved	achieve	VERB
cana-5407	25	10	in	in	ADP
cana-5407	25	11	the	the	DET
cana-5407	25	12	field	field	NOUN
cana-5407	25	13	of	of	ADP
cana-5407	25	14	voice	voice	NOUN
cana-5407	25	15	conversion	conversion	NOUN
cana-5407	25	16	,	,	PUNCT
cana-5407	25	17	there	there	PRON
cana-5407	25	18	are	be	VERB
cana-5407	25	19	still	still	ADV
cana-5407	25	20	a	a	DET
cana-5407	25	21	number	number	NOUN
cana-5407	25	22	of	of	ADP
cana-5407	25	23	research	research	NOUN
cana-5407	25	24	gaps	gap	NOUN
cana-5407	25	25	to	to	PART
cana-5407	25	26	be	be	AUX
cana-5407	25	27	tackled	tackle	VERB
cana-5407	25	28	[	[	X
cana-5407	25	29	7	7	NUM
cana-5407	25	30	]	]	PUNCT
cana-5407	25	31	.	.	PUNCT
cana-5407	26	1	for	for	ADP
cana-5407	26	2	instance	instance	NOUN
cana-5407	26	3	,	,	PUNCT
cana-5407	26	4	most	most	ADV
cana-5407	26	5	existent	existent	ADJ
cana-5407	26	6	methods	method	NOUN
cana-5407	26	7	bear	bear	VERB
cana-5407	26	8	a	a	DET
cana-5407	26	9	number	number	NOUN
cana-5407	26	10	of	of	ADP
cana-5407	26	11	difficulties	difficulty	NOUN
cana-5407	26	12	with	with	ADP
cana-5407	26	13	large	large	ADJ
cana-5407	26	14	volumes	volume	NOUN
cana-5407	26	15	of	of	ADP
cana-5407	26	16	data	datum	NOUN
cana-5407	26	17	processing	processing	NOUN
cana-5407	26	18	.	.	PUNCT
cana-5407	27	1	therefore	therefore	ADV
cana-5407	27	2	,	,	PUNCT
cana-5407	27	3	they	they	PRON
cana-5407	27	4	are	be	AUX
cana-5407	27	5	not	not	PART
cana-5407	27	6	fully	fully	ADV
cana-5407	27	7	applicable	applicable	ADJ
cana-5407	27	8	for	for	ADP
cana-5407	27	9	a	a	DET
cana-5407	27	10	titanic	titanic	ADJ
cana-5407	27	11	array	array	NOUN
cana-5407	27	12	of	of	ADP
cana-5407	27	13	tasks	task	NOUN
cana-5407	27	14	.	.	PUNCT
cana-5407	28	1	additionally	additionally	ADV
cana-5407	28	2	,	,	PUNCT
cana-5407	28	3	the	the	DET
cana-5407	28	4	emotional	emotional	ADJ
cana-5407	28	5	nuances	nuance	NOUN
cana-5407	28	6	and	and	CCONJ
cana-5407	28	7	subtle	subtle	ADJ
cana-5407	28	8	details	detail	NOUN
cana-5407	28	9	of	of	ADP
cana-5407	28	10	the	the	DET
cana-5407	28	11	original	original	ADJ
cana-5407	28	12	voice	voice	NOUN
cana-5407	28	13	are	be	AUX
cana-5407	28	14	particularly	particularly	ADV
cana-5407	28	15	difficult	difficult	ADJ
cana-5407	28	16	to	to	PART
cana-5407	28	17	preserve	preserve	VERB
cana-5407	28	18	.	.	PUNCT
cana-5407	29	1	these	these	DET
cana-5407	29	2	factors	factor	NOUN
cana-5407	29	3	deteriorate	deteriorate	VERB
cana-5407	29	4	the	the	DET
cana-5407	29	5	authenticity	authenticity	NOUN
cana-5407	29	6	and	and	CCONJ
cana-5407	29	7	the	the	DET
cana-5407	29	8	effectiveness	effectiveness	NOUN
cana-5407	29	9	of	of	ADP
cana-5407	29	10	the	the	DET
cana-5407	29	11	converted	converted	ADJ
cana-5407	29	12	voice	voice	NOUN
cana-5407	29	13	[	[	X
cana-5407	29	14	8	8	NUM
cana-5407	29	15	]	]	PUNCT
cana-5407	29	16	.	.	PUNCT
cana-5407	30	1	therefore	therefore	ADV
cana-5407	30	2	,	,	PUNCT
cana-5407	30	3	understanding	understand	VERB
cana-5407	30	4	and	and	CCONJ
cana-5407	30	5	removing	remove	VERB
cana-5407	30	6	these	these	DET
cana-5407	30	7	limitations	limitation	NOUN
cana-5407	30	8	may	may	AUX
cana-5407	30	9	contribute	contribute	VERB
cana-5407	30	10	to	to	ADP
cana-5407	30	11	improving	improve	VERB
cana-5407	30	12	the	the	DET
cana-5407	30	13	capacities	capacity	NOUN
cana-5407	30	14	and	and	CCONJ
cana-5407	30	15	applicability	applicability	NOUN
cana-5407	30	16	of	of	ADP
cana-5407	30	17	speech	speech	NOUN
cana-5407	30	18	conversion	conversion	NOUN
cana-5407	30	19	technologies	technology	NOUN
cana-5407	30	20	.	.	PUNCT
cana-5407	31	1	hence	hence	ADV
cana-5407	31	2	,	,	PUNCT
cana-5407	31	3	they	they	PRON
cana-5407	31	4	may	may	AUX
cana-5407	31	5	serve	serve	VERB
cana-5407	31	6	as	as	ADP
cana-5407	31	7	prospective	prospective	ADJ
cana-5407	31	8	directions	direction	NOUN
cana-5407	31	9	for	for	ADP
cana-5407	31	10	subsequent	subsequent	ADJ
cana-5407	31	11	research	research	NOUN
cana-5407	31	12	.	.	PUNCT
cana-5407	32	1	literature	literature	PROPN
cana-5407	32	2	review	review	PROPN
cana-5407	32	3	:	:	PUNCT
cana-5407	32	4	berrak	berrak	NOUN
cana-5407	32	5	sisman	sisman	NOUN
cana-5407	32	6	et	et	PROPN
cana-5407	32	7	al	al	PROPN
cana-5407	33	1	[	[	X
cana-5407	33	2	9	9	NUM
cana-5407	33	3	]	]	PUNCT
cana-5407	33	4	provided	provide	VERB
cana-5407	33	5	a	a	DET
cana-5407	33	6	thorough	thorough	ADJ
cana-5407	33	7	analysis	analysis	NOUN
cana-5407	33	8	of	of	ADP
cana-5407	33	9	the	the	DET
cana-5407	33	10	current	current	ADJ
cana-5407	33	11	state	state	NOUN
cana-5407	33	12	-	-	PUNCT
cana-5407	33	13	of	of	ADP
cana-5407	33	14	-	-	PUNCT
cana-5407	33	15	the	the	DET
cana-5407	33	16	-	-	PUNCT
cana-5407	33	17	art	art	NOUN
cana-5407	33	18	speech	speech	NOUN
cana-5407	33	19	conversion	conversion	NOUN
cana-5407	33	20	techniques	technique	NOUN
cana-5407	33	21	and	and	CCONJ
cana-5407	33	22	their	their	PRON
cana-5407	33	23	manner	manner	NOUN
cana-5407	33	24	of	of	ADP
cana-5407	33	25	evaluating	evaluate	VERB
cana-5407	33	26	performance	performance	NOUN
cana-5407	33	27	.	.	PUNCT
cana-5407	34	1	it	it	PRON
cana-5407	34	2	includes	include	VERB
cana-5407	34	3	various	various	ADJ
cana-5407	34	4	approaches	approach	NOUN
cana-5407	34	5	,	,	PUNCT
cana-5407	34	6	ranging	range	VERB
cana-5407	34	7	from	from	ADP
cana-5407	34	8	statistical	statistical	ADJ
cana-5407	34	9	methods	method	NOUN
cana-5407	34	10	to	to	ADP
cana-5407	34	11	deep	deep	ADJ
cana-5407	34	12	learning	learning	NOUN
cana-5407	34	13	.	.	PUNCT
cana-5407	35	1	the	the	DET
cana-5407	35	2	authors	author	NOUN
cana-5407	35	3	have	have	AUX
cana-5407	35	4	considered	consider	VERB
cana-5407	35	5	the	the	DET
cana-5407	35	6	potential	potential	ADJ
cana-5407	35	7	benefits	benefit	NOUN
cana-5407	35	8	and	and	CCONJ
cana-5407	35	9	drawbacks	drawback	NOUN
cana-5407	35	10	of	of	ADP
cana-5407	35	11	these	these	DET
cana-5407	35	12	techniques	technique	NOUN
cana-5407	35	13	.	.	PUNCT
cana-5407	36	1	furthermore	furthermore	ADV
cana-5407	36	2	,	,	PUNCT
cana-5407	36	3	they	they	PRON
cana-5407	36	4	will	will	AUX
cana-5407	36	5	go	go	VERB
cana-5407	36	6	over	over	ADP
cana-5407	36	7	the	the	DET
cana-5407	36	8	latest	late	ADJ
cana-5407	36	9	voice	voice	NOUN
cana-5407	36	10	conversion	conversion	NOUN
cana-5407	36	11	challenges	challenge	NOUN
cana-5407	36	12	.	.	PUNCT
cana-5407	37	1	voice	voice	NOUN
cana-5407	37	2	conversion	conversion	NOUN
cana-5407	37	3	involves	involve	VERB
cana-5407	37	4	various	various	ADJ
cana-5407	37	5	speech	speech	NOUN
cana-5407	37	6	processing	processing	NOUN
cana-5407	37	7	techniques	technique	NOUN
cana-5407	37	8	including	include	VERB
cana-5407	37	9	voice	voice	NOUN
cana-5407	37	10	analysis	analysis	NOUN
cana-5407	37	11	,	,	PUNCT
cana-5407	37	12	spectrum	spectrum	NOUN
cana-5407	37	13	conversion	conversion	NOUN
cana-5407	37	14	,	,	PUNCT
cana-5407	37	15	prosody	prosody	NOUN
cana-5407	37	16	conversion	conversion	NOUN
cana-5407	37	17	,	,	PUNCT
cana-5407	37	18	speaker	speaker	NOUN
cana-5407	37	19	characterisation	characterisation	NOUN
cana-5407	37	20	and	and	CCONJ
cana-5407	37	21	vocoding	vocoding	NOUN
cana-5407	37	22	.	.	PUNCT
cana-5407	38	1	briefly	briefly	ADV
cana-5407	38	2	,	,	PUNCT
cana-5407	38	3	thanks	thank	NOUN
cana-5407	38	4	to	to	ADP
cana-5407	38	5	recent	recent	ADJ
cana-5407	38	6	advances	advance	NOUN
cana-5407	38	7	in	in	ADP
cana-5407	38	8	both	both	CCONJ
cana-5407	38	9	the	the	DET
cana-5407	38	10	theory	theory	NOUN
cana-5407	38	11	and	and	CCONJ
cana-5407	38	12	practice	practice	NOUN
cana-5407	38	13	,	,	PUNCT
cana-5407	38	14	they	they	PRON
cana-5407	38	15	can	can	AUX
cana-5407	38	16	generate	generate	VERB
cana-5407	38	17	voice	voice	NOUN
cana-5407	38	18	quality	quality	NOUN
cana-5407	38	19	capable	capable	ADJ
cana-5407	38	20	of	of	ADP
cana-5407	38	21	perfectly	perfectly	ADV
cana-5407	38	22	imitating	imitate	VERB
cana-5407	38	23	human	human	ADJ
cana-5407	38	24	speech	speech	NOUN
cana-5407	38	25	and	and	CCONJ
cana-5407	38	26	the	the	DET
cana-5407	38	27	speaker	speaker	NOUN
cana-5407	38	28	.	.	PUNCT
cana-5407	39	1	yi	yi	PROPN
cana-5407	40	1	zhao	zhao	PROPN
cana-5407	40	2	et	et	PROPN
cana-5407	40	3	al	al	PROPN
cana-5407	41	1	[	[	X
cana-5407	41	2	10	10	NUM
cana-5407	41	3	]	]	PUNCT
cana-5407	41	4	planned	plan	VERB
cana-5407	41	5	and	and	CCONJ
cana-5407	41	6	executed	execute	VERB
cana-5407	41	7	the	the	DET
cana-5407	41	8	third	third	ADJ
cana-5407	41	9	edition	edition	NOUN
cana-5407	41	10	of	of	ADP
cana-5407	41	11	the	the	DET
cana-5407	41	12	challenge	challenge	NOUN
cana-5407	41	13	,	,	PUNCT
cana-5407	41	14	and	and	CCONJ
cana-5407	41	15	developed	develop	VERB
cana-5407	41	16	and	and	CCONJ
cana-5407	41	17	disseminated	disseminate	VERB
cana-5407	41	18	a	a	DET
cana-5407	41	19	new	new	ADJ
cana-5407	41	20	database	database	NOUN
cana-5407	41	21	for	for	ADP
cana-5407	41	22	two	two	NUM
cana-5407	41	23	specific	specific	ADJ
cana-5407	41	24	tasks	task	NOUN
cana-5407	41	25	,	,	PUNCT
cana-5407	41	26	intra	intra	ADJ
cana-5407	41	27	-	-	ADJ
cana-5407	41	28	lingual	lingual	ADJ
cana-5407	41	29	semi	semi	NOUN
cana-5407	41	30	-	-	NOUN
cana-5407	41	31	parallal	parallal	NOUN
cana-5407	41	32	and	and	CCONJ
cana-5407	41	33	cross	cross	NOUN
cana-5407	41	34	-	-	NOUN
cana-5407	41	35	lingual	lingual	ADJ
cana-5407	41	36	vc	vc	NOUN
cana-5407	41	37	.	.	PUNCT
cana-5407	42	1	after	after	ADP
cana-5407	42	2	a	a	DET
cana-5407	42	3	two	two	NUM
cana-5407	42	4	-	-	PUNCT
cana-5407	42	5	month	month	NOUN
cana-5407	42	6	challenge	challenge	NOUN
cana-5407	42	7	period	period	NOUN
cana-5407	42	8	,	,	PUNCT
cana-5407	42	9	the	the	DET
cana-5407	42	10	authors	author	NOUN
cana-5407	42	11	have	have	AUX
cana-5407	42	12	been	be	AUX
cana-5407	42	13	able	able	ADJ
cana-5407	42	14	to	to	PART
cana-5407	42	15	collect	collect	VERB
cana-5407	42	16	a	a	DET
cana-5407	42	17	total	total	NOUN
cana-5407	42	18	of	of	ADP
cana-5407	42	19	33	33	NUM
cana-5407	42	20	entries	entry	NOUN
cana-5407	42	21	,	,	PUNCT
cana-5407	42	22	including	include	VERB
cana-5407	42	23	three	three	NUM
cana-5407	42	24	baseline	baseline	NOUN
cana-5407	42	25	submissions	submission	NOUN
cana-5407	42	26	made	make	VERB
cana-5407	42	27	using	use	VERB
cana-5407	42	28	the	the	DET
cana-5407	42	29	database	database	NOUN
cana-5407	42	30	.	.	PUNCT
cana-5407	43	1	it	it	PRON
cana-5407	43	2	has	have	AUX
cana-5407	43	3	been	be	AUX
cana-5407	43	4	observed	observe	VERB
cana-5407	43	5	that	that	SCONJ
cana-5407	43	6	the	the	DET
cana-5407	43	7	progress	progress	NOUN
cana-5407	43	8	of	of	ADP
cana-5407	43	9	vc	vc	PROPN
cana-5407	43	10	techniques	technique	NOUN
cana-5407	43	11	improves	improve	VERB
cana-5407	43	12	drastically	drastically	ADV
cana-5407	43	13	as	as	ADP
cana-5407	43	14	a	a	DET
cana-5407	43	15	result	result	NOUN
cana-5407	43	16	of	of	ADP
cana-5407	43	17	higher	high	ADJ
cana-5407	43	18	quality	quality	NOUN
cana-5407	43	19	deep	deep	ADJ
cana-5407	43	20	learning	learning	NOUN
cana-5407	43	21	techniques	technique	NOUN
cana-5407	43	22	.	.	PUNCT
cana-5407	44	1	for	for	ADP
cana-5407	44	2	instance	instance	NOUN
cana-5407	44	3	,	,	PUNCT
cana-5407	44	4	in	in	ADP
cana-5407	44	5	the	the	DET
cana-5407	44	6	task	task	NOUN
cana-5407	44	7	of	of	ADP
cana-5407	44	8	intra	intra	ADJ
cana-5407	44	9	-	-	ADJ
cana-5407	44	10	lingual	lingual	ADJ
cana-5407	44	11	semi	semi	ADJ
cana-5407	44	12	-	-	ADJ
cana-5407	44	13	parallel	parallel	ADJ
cana-5407	44	14	vc	vc	NOUN
cana-5407	44	15	,	,	PUNCT
cana-5407	44	16	many	many	ADJ
cana-5407	44	17	of	of	ADP
cana-5407	44	18	the	the	DET
cana-5407	44	19	systems	system	NOUN
cana-5407	44	20	’	'	PUNCT
cana-5407	44	21	speaker	speaker	NOUN
cana-5407	44	22	similarity	similarity	NOUN
cana-5407	44	23	scores	score	NOUN
cana-5407	44	24	were	be	AUX
cana-5407	44	25	within	within	ADP
cana-5407	44	26	the	the	DET
cana-5407	44	27	range	range	NOUN
cana-5407	44	28	of	of	ADP
cana-5407	44	29	those	those	PRON
cana-5407	44	30	of	of	ADP
cana-5407	44	31	the	the	DET
cana-5407	44	32	target	target	NOUN
cana-5407	44	33	speakers	speaker	NOUN
cana-5407	44	34	.	.	PUNCT
cana-5407	45	1	nevertheless	nevertheless	ADV
cana-5407	45	2	,	,	PUNCT
cana-5407	45	3	they	they	PRON
cana-5407	45	4	have	have	AUX
cana-5407	45	5	ascertained	ascertain	VERB
cana-5407	45	6	that	that	SCONJ
cana-5407	45	7	none	none	NOUN
cana-5407	45	8	of	of	ADP
cana-5407	45	9	them	they	PRON
cana-5407	45	10	has	have	AUX
cana-5407	45	11	achieved	achieve	VERB
cana-5407	45	12	the	the	DET
cana-5407	45	13	degree	degree	NOUN
cana-5407	45	14	of	of	ADP
cana-5407	45	15	naturalness	naturalness	NOUN
cana-5407	45	16	that	that	PRON
cana-5407	45	17	humans	human	NOUN
cana-5407	45	18	exhibit	exhibit	VERB
cana-5407	45	19	for	for	ADP
cana-5407	45	20	the	the	DET
cana-5407	45	21	same	same	ADJ
cana-5407	45	22	job	job	NOUN
cana-5407	45	23	.	.	PUNCT
cana-5407	46	1	the	the	DET
cana-5407	46	2	crosslingual	crosslingual	ADJ
cana-5407	46	3	conversion	conversion	NOUN
cana-5407	46	4	proved	prove	VERB
cana-5407	46	5	to	to	PART
cana-5407	46	6	be	be	AUX
cana-5407	46	7	more	more	ADV
cana-5407	46	8	difficult	difficult	ADJ
cana-5407	46	9	,	,	PUNCT
cana-5407	46	10	as	as	SCONJ
cana-5407	46	11	all	all	DET
cana-5407	46	12	systems	system	NOUN
cana-5407	46	13	obtained	obtain	VERB
cana-5407	46	14	relatively	relatively	ADV
cana-5407	46	15	lower	low	ADJ
cana-5407	46	16	naturalness	naturalness	NOUN
cana-5407	46	17	and	and	CCONJ
cana-5407	46	18	similarity	similarity	NOUN
cana-5407	46	19	column	column	NOUN
cana-5407	46	20	heading	head	VERB
cana-5407	46	21	ratings	rating	NOUN
cana-5407	46	22	compared	compare	VERB
cana-5407	46	23	to	to	ADP
cana-5407	46	24	the	the	DET
cana-5407	46	25	intra	intra	ADJ
cana-5407	46	26	-	-	ADJ
cana-5407	46	27	lingual	lingual	ADJ
cana-5407	46	28	conversion	conversion	NOUN
cana-5407	46	29	task	task	NOUN
cana-5407	46	30	.	.	PUNCT
cana-5407	47	1	kun	kun	PROPN
cana-5407	47	2	zhou	zhou	PROPN
cana-5407	47	3	et	et	PROPN
cana-5407	47	4	al	al	PROPN
cana-5407	48	1	[	[	X
cana-5407	48	2	11	11	NUM
cana-5407	48	3	]	]	PUNCT
cana-5407	48	4	introduced	introduce	VERB
cana-5407	48	5	a	a	DET
cana-5407	48	6	review	review	NOUN
cana-5407	48	7	of	of	ADP
cana-5407	48	8	the	the	DET
cana-5407	48	9	most	most	ADV
cana-5407	48	10	innovative	innovative	ADJ
cana-5407	48	11	emotional	emotional	ADJ
cana-5407	48	12	voice	voice	NOUN
cana-5407	48	13	conversion	conversion	NOUN
cana-5407	48	14	research	research	NOUN
cana-5407	48	15	and	and	CCONJ
cana-5407	48	16	available	available	ADJ
cana-5407	48	17	emotional	emotional	ADJ
cana-5407	48	18	speech	speech	NOUN
cana-5407	48	19	databases	database	NOUN
cana-5407	48	20	.	.	PUNCT
cana-5407	49	1	then	then	ADV
cana-5407	49	2	,	,	PUNCT
cana-5407	49	3	the	the	DET
cana-5407	49	4	authors	author	NOUN
cana-5407	49	5	argued	argue	VERB
cana-5407	49	6	the	the	DET
cana-5407	49	7	necessity	necessity	NOUN
cana-5407	49	8	to	to	PART
cana-5407	49	9	create	create	VERB
cana-5407	49	10	an	an	DET
cana-5407	49	11	emotional	emotional	ADJ
cana-5407	49	12	speech	speech	NOUN
cana-5407	49	13	database	database	NOUN
cana-5407	49	14	for	for	ADP
cana-5407	49	15	the	the	DET
cana-5407	49	16	first	first	ADJ
cana-5407	49	17	time	time	NOUN
cana-5407	49	18	,	,	PUNCT
cana-5407	49	19	addressing	address	VERB
cana-5407	49	20	the	the	DET
cana-5407	49	21	emerging	emerge	VERB
cana-5407	49	22	requirements	requirement	NOUN
cana-5407	49	23	at	at	ADP
cana-5407	49	24	a	a	DET
cana-5407	49	25	rising	rise	VERB
cana-5407	49	26	trend	trend	NOUN
cana-5407	49	27	communications	communication	NOUN
cana-5407	49	28	on	on	ADP
cana-5407	49	29	applied	apply	VERB
cana-5407	49	30	nonlinear	nonlinear	ADJ
cana-5407	49	31	analysis	analysis	NOUN
cana-5407	49	32	issn	issn	NOUN
cana-5407	49	33	:	:	PUNCT
cana-5407	49	34	1074	1074	NUM
cana-5407	49	35	-	-	PUNCT
cana-5407	49	36	133x	133x	NUM
cana-5407	49	37	vol	vol	VERB
cana-5407	49	38	32	32	NUM
cana-5407	49	39	no	no	NOUN
cana-5407	49	40	.	.	PUNCT
cana-5407	50	1	10s	10	NOUN
cana-5407	50	2	(	(	PUNCT
cana-5407	50	3	2025	2025	NUM
cana-5407	50	4	)	)	PUNCT
cana-5407	50	5	2174	2174	NUM
cana-5407	50	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-5407	50	7	in	in	ADP
cana-5407	50	8	research	research	NOUN
cana-5407	50	9	.	.	PUNCT
cana-5407	51	1	the	the	DET
cana-5407	51	2	work	work	NOUN
cana-5407	51	3	has	have	AUX
cana-5407	51	4	released	release	VERB
cana-5407	51	5	the	the	DET
cana-5407	51	6	emotional	emotional	ADJ
cana-5407	51	7	speech	speech	NOUN
cana-5407	51	8	database	database	NOUN
cana-5407	51	9	esd	esd	VERB
cana-5407	51	10	to	to	ADP
cana-5407	51	11	the	the	DET
cana-5407	51	12	entire	entire	ADJ
cana-5407	51	13	research	research	NOUN
cana-5407	51	14	community	community	NOUN
cana-5407	51	15	.	.	PUNCT
cana-5407	52	1	in	in	ADP
cana-5407	52	2	brief	brief	NOUN
cana-5407	52	3	,	,	PUNCT
cana-5407	52	4	esd	esd	PROPN
cana-5407	52	5	contains	contain	VERB
cana-5407	52	6	350	350	NUM
cana-5407	52	7	parallel	parallel	ADJ
cana-5407	52	8	utterances	utterance	NOUN
cana-5407	52	9	for	for	ADP
cana-5407	52	10	building	build	VERB
cana-5407	52	11	the	the	DET
cana-5407	52	12	model	model	NOUN
cana-5407	52	13	and	and	CCONJ
cana-5407	52	14	converting	convert	VERB
cana-5407	52	15	10	10	NUM
cana-5407	52	16	speakers	speaker	NOUN
cana-5407	52	17	from	from	ADP
cana-5407	52	18	two	two	NUM
cana-5407	52	19	native	native	ADJ
cana-5407	52	20	speakers	speaker	NOUN
cana-5407	52	21	for	for	ADP
cana-5407	52	22	english	english	PROPN
cana-5407	52	23	and	and	CCONJ
cana-5407	52	24	five	five	NUM
cana-5407	52	25	emotion	emotion	NOUN
cana-5407	52	26	categories	category	NOUN
cana-5407	52	27	of	of	ADP
cana-5407	52	28	chinese	chinese	PROPN
cana-5407	52	29	.	.	PUNCT
cana-5407	53	1	a	a	DET
cana-5407	53	2	total	total	NOUN
cana-5407	53	3	of	of	ADP
cana-5407	53	4	almost	almost	ADV
cana-5407	53	5	29	29	NUM
cana-5407	53	6	hours	hour	NOUN
cana-5407	53	7	of	of	ADP
cana-5407	53	8	speech	speech	NOUN
cana-5407	53	9	data	datum	NOUN
cana-5407	53	10	were	be	AUX
cana-5407	53	11	used	use	VERB
cana-5407	53	12	in	in	ADP
cana-5407	53	13	a	a	DET
cana-5407	53	14	strictly	strictly	ADV
cana-5407	53	15	controlled	control	VERB
cana-5407	53	16	acoustic	acoustic	ADJ
cana-5407	53	17	environment	environment	NOUN
cana-5407	53	18	.	.	PUNCT
cana-5407	54	1	the	the	DET
cana-5407	54	2	database	database	NOUN
cana-5407	54	3	is	be	AUX
cana-5407	54	4	adequate	adequate	ADJ
cana-5407	54	5	for	for	ADP
cana-5407	54	6	multi	multi	ADJ
cana-5407	54	7	-	-	NOUN
cana-5407	54	8	speaker	speaker	NOUN
cana-5407	54	9	and	and	CCONJ
cana-5407	54	10	multi	multi	ADJ
cana-5407	54	11	-	-	ADJ
cana-5407	54	12	system	system	ADJ
cana-5407	54	13	investigations	investigation	NOUN
cana-5407	54	14	on	on	ADP
cana-5407	54	15	emotional	emotional	ADJ
cana-5407	54	16	voice	voice	NOUN
cana-5407	54	17	conversion	conversion	NOUN
cana-5407	54	18	in	in	ADP
cana-5407	54	19	two	two	NUM
cana-5407	54	20	different	different	ADJ
cana-5407	54	21	languages	language	NOUN
cana-5407	54	22	.	.	PUNCT
cana-5407	55	1	disong	disong	PROPN
cana-5407	55	2	wang	wang	PROPN
cana-5407	55	3	et	et	PROPN
cana-5407	55	4	al	al	PROPN
cana-5407	56	1	[	[	X
cana-5407	56	2	12	12	NUM
cana-5407	56	3	]	]	X
cana-5407	56	4	utilized	utilize	VERB
cana-5407	56	5	technique	technique	NOUN
cana-5407	56	6	of	of	ADP
cana-5407	56	7	vector	vector	NOUN
cana-5407	56	8	quantization	quantization	NOUN
cana-5407	56	9	(	(	PUNCT
cana-5407	56	10	vq	vq	NOUN
cana-5407	56	11	)	)	PUNCT
cana-5407	56	12	for	for	ADP
cana-5407	56	13	content	content	NOUN
cana-5407	56	14	encoding	encoding	NOUN
cana-5407	56	15	,	,	PUNCT
cana-5407	56	16	while	while	SCONJ
cana-5407	56	17	mutual	mutual	ADJ
cana-5407	56	18	information	information	NOUN
cana-5407	56	19	(	(	PUNCT
cana-5407	56	20	mi	mi	NOUN
cana-5407	56	21	)	)	PUNCT
cana-5407	56	22	is	be	AUX
cana-5407	56	23	introduced	introduce	VERB
cana-5407	56	24	as	as	ADP
cana-5407	56	25	the	the	DET
cana-5407	56	26	correlation	correlation	NOUN
cana-5407	56	27	measure	measure	NOUN
cana-5407	56	28	during	during	ADP
cana-5407	56	29	training	training	NOUN
cana-5407	56	30	.	.	PUNCT
cana-5407	57	1	this	this	DET
cana-5407	57	2	approach	approach	NOUN
cana-5407	57	3	strives	strive	VERB
cana-5407	57	4	for	for	ADP
cana-5407	57	5	effective	effective	ADJ
cana-5407	57	6	disentanglement	disentanglement	NOUN
cana-5407	57	7	of	of	ADP
cana-5407	57	8	content	content	NOUN
cana-5407	57	9	,	,	PUNCT
cana-5407	57	10	speaker	speaker	NOUN
cana-5407	57	11	,	,	PUNCT
cana-5407	57	12	and	and	CCONJ
cana-5407	57	13	pitch	pitch	NOUN
cana-5407	57	14	representations	representation	NOUN
cana-5407	57	15	by	by	ADP
cana-5407	57	16	minimizing	minimize	VERB
cana-5407	57	17	their	their	PRON
cana-5407	57	18	inter	inter	NOUN
cana-5407	57	19	-	-	NOUN
cana-5407	57	20	dependencies	dependency	NOUN
cana-5407	57	21	in	in	ADP
cana-5407	57	22	an	an	DET
cana-5407	57	23	unsupervised	unsupervised	ADJ
cana-5407	57	24	way	way	NOUN
cana-5407	57	25	.	.	PUNCT
cana-5407	58	1	yinghao	yinghao	ADJ
cana-5407	58	2	aaron	aaron	PROPN
cana-5407	58	3	li	li	PROPN
cana-5407	59	1	[	[	X
cana-5407	59	2	13	13	NUM
cana-5407	59	3	]	]	PUNCT
cana-5407	59	4	introduced	introduce	VERB
cana-5407	59	5	an	an	DET
cana-5407	59	6	unsupervised	unsupervised	ADJ
cana-5407	59	7	non	non	ADJ
cana-5407	59	8	-	-	ADJ
cana-5407	59	9	parallel	parallel	ADJ
cana-5407	59	10	many	many	ADJ
cana-5407	59	11	-	-	PUNCT
cana-5407	59	12	to	to	ADP
cana-5407	59	13	-	-	PUNCT
cana-5407	59	14	many	many	ADJ
cana-5407	59	15	voice	voice	NOUN
cana-5407	59	16	conversion	conversion	NOUN
cana-5407	59	17	(	(	PUNCT
cana-5407	59	18	vc	vc	NOUN
cana-5407	59	19	)	)	PUNCT
cana-5407	59	20	method	method	NOUN
cana-5407	59	21	using	use	VERB
cana-5407	59	22	a	a	DET
cana-5407	59	23	generative	generative	ADJ
cana-5407	59	24	adversarial	adversarial	ADJ
cana-5407	59	25	network	network	NOUN
cana-5407	59	26	(	(	PUNCT
cana-5407	59	27	gan	gan	PROPN
cana-5407	59	28	)	)	PUNCT
cana-5407	59	29	called	call	VERB
cana-5407	59	30	stargan	stargan	PROPN
cana-5407	59	31	v2	v2	PROPN
cana-5407	59	32	.	.	PUNCT
cana-5407	60	1	the	the	DET
cana-5407	60	2	model	model	NOUN
cana-5407	60	3	substantially	substantially	ADV
cana-5407	60	4	outperforms	outperform	VERB
cana-5407	60	5	the	the	DET
cana-5407	60	6	past	past	ADJ
cana-5407	60	7	vc	vc	PROPN
cana-5407	60	8	models	model	NOUN
cana-5407	60	9	by	by	ADP
cana-5407	60	10	using	use	VERB
cana-5407	60	11	a	a	DET
cana-5407	60	12	balanced	balanced	ADJ
cana-5407	60	13	mix	mix	NOUN
cana-5407	60	14	of	of	ADP
cana-5407	60	15	adversarial	adversarial	ADJ
cana-5407	60	16	source	source	NOUN
cana-5407	60	17	classifier	classifier	NOUN
cana-5407	60	18	loss	loss	NOUN
cana-5407	60	19	and	and	CCONJ
cana-5407	60	20	perceptual	perceptual	ADJ
cana-5407	60	21	loss	loss	NOUN
cana-5407	60	22	.	.	PUNCT
cana-5407	61	1	with	with	ADP
cana-5407	61	2	only	only	ADV
cana-5407	61	3	training	train	VERB
cana-5407	61	4	a	a	DET
cana-5407	61	5	model	model	NOUN
cana-5407	61	6	solely	solely	ADV
cana-5407	61	7	based	base	VERB
cana-5407	61	8	on	on	ADP
cana-5407	61	9	20	20	NUM
cana-5407	61	10	english	english	ADJ
cana-5407	61	11	speakers	speaker	NOUN
cana-5407	61	12	,	,	PUNCT
cana-5407	61	13	the	the	DET
cana-5407	61	14	model	model	NOUN
cana-5407	61	15	can	can	AUX
cana-5407	61	16	be	be	AUX
cana-5407	61	17	easily	easily	ADV
cana-5407	61	18	adapted	adapt	VERB
cana-5407	61	19	to	to	ADP
cana-5407	61	20	a	a	DET
cana-5407	61	21	wide	wide	ADJ
cana-5407	61	22	range	range	NOUN
cana-5407	61	23	of	of	ADP
cana-5407	61	24	voice	voice	NOUN
cana-5407	61	25	conversion	conversion	NOUN
cana-5407	61	26	tasks	task	NOUN
cana-5407	61	27	such	such	ADJ
cana-5407	61	28	as	as	ADP
cana-5407	61	29	any	any	DET
cana-5407	61	30	-	-	PUNCT
cana-5407	61	31	to	to	ADP
cana-5407	61	32	-	-	PUNCT
cana-5407	61	33	many	many	ADJ
cana-5407	61	34	,	,	PUNCT
cana-5407	61	35	cross	cross	NOUN
cana-5407	61	36	-	-	ADJ
cana-5407	61	37	lingual	lingual	ADJ
cana-5407	61	38	,	,	PUNCT
cana-5407	61	39	singing	singe	VERB
cana-5407	61	40	conversion	conversion	NOUN
cana-5407	61	41	.	.	PUNCT
cana-5407	62	1	the	the	DET
cana-5407	62	2	framework	framework	NOUN
cana-5407	62	3	can	can	AUX
cana-5407	62	4	also	also	ADV
cana-5407	62	5	convert	convert	VERB
cana-5407	62	6	plain	plain	ADJ
cana-5407	62	7	reading	reading	NOUN
cana-5407	62	8	speech	speech	NOUN
cana-5407	62	9	to	to	ADP
cana-5407	62	10	their	their	PRON
cana-5407	62	11	corresponding	corresponding	ADJ
cana-5407	62	12	stylistic	stylistic	ADJ
cana-5407	62	13	speech	speech	NOUN
cana-5407	62	14	,	,	PUNCT
cana-5407	62	15	such	such	ADJ
cana-5407	62	16	as	as	ADP
cana-5407	62	17	emotional	emotional	ADJ
cana-5407	62	18	,	,	PUNCT
cana-5407	62	19	and	and	CCONJ
cana-5407	62	20	falsetto	falsetto	NOUN
cana-5407	62	21	.	.	PUNCT
cana-5407	63	1	songxiang	songxiang	PROPN
cana-5407	63	2	liu	liu	PROPN
cana-5407	63	3	et	et	PROPN
cana-5407	63	4	al	al	PROPN
cana-5407	64	1	[	[	X
cana-5407	64	2	14	14	NUM
cana-5407	64	3	]	]	PUNCT
cana-5407	64	4	proposed	propose	VERB
cana-5407	64	5	an	an	DET
cana-5407	64	6	entirely	entirely	ADV
cana-5407	64	7	novel	novel	ADJ
cana-5407	64	8	voice	voice	NOUN
cana-5407	64	9	conversion	conversion	NOUN
cana-5407	64	10	technique	technique	NOUN
cana-5407	64	11	involving	involve	VERB
cana-5407	64	12	an	an	DET
cana-5407	64	13	arbitrary	arbitrary	ADJ
cana-5407	64	14	number	number	NOUN
cana-5407	64	15	of	of	ADP
cana-5407	64	16	input	input	NOUN
cana-5407	64	17	and	and	CCONJ
cana-5407	64	18	output	output	NOUN
cana-5407	64	19	locations	location	NOUN
cana-5407	64	20	,	,	PUNCT
cana-5407	64	21	a	a	DET
cana-5407	64	22	sequence	sequence	NOUN
cana-5407	64	23	,	,	PUNCT
cana-5407	64	24	and	and	CCONJ
cana-5407	64	25	even	even	ADV
cana-5407	64	26	non	non	ADJ
cana-5407	64	27	-	-	ADJ
cana-5407	64	28	parallel	parallel	ADJ
cana-5407	64	29	transmission	transmission	NOUN
cana-5407	64	30	data	datum	NOUN
cana-5407	64	31	.	.	PUNCT
cana-5407	65	1	the	the	DET
cana-5407	65	2	system	system	NOUN
cana-5407	65	3	’s	’s	PART
cana-5407	65	4	training	training	NOUN
cana-5407	65	5	is	be	AUX
cana-5407	65	6	guided	guide	VERB
cana-5407	65	7	by	by	ADP
cana-5407	65	8	voice	voice	NOUN
cana-5407	65	9	-	-	PUNCT
cana-5407	65	10	based	base	VERB
cana-5407	65	11	textual	textual	ADJ
cana-5407	65	12	information	information	NOUN
cana-5407	65	13	.	.	PUNCT
cana-5407	66	1	while	while	SCONJ
cana-5407	66	2	training	train	VERB
cana-5407	66	3	the	the	DET
cana-5407	66	4	phoneme	phoneme	NOUN
cana-5407	66	5	recognizer	recognizer	NOUN
cana-5407	66	6	,	,	PUNCT
cana-5407	66	7	the	the	DET
cana-5407	66	8	authors	author	NOUN
cana-5407	66	9	combined	combine	VERB
cana-5407	66	10	the	the	DET
cana-5407	66	11	middle	middle	ADJ
cana-5407	66	12	layer	layer	NOUN
cana-5407	66	13	of	of	ADP
cana-5407	66	14	the	the	DET
cana-5407	66	15	encoder	encoder	NOUN
cana-5407	66	16	with	with	ADP
cana-5407	66	17	the	the	DET
cana-5407	66	18	bne	bne	PROPN
cana-5407	66	19	and	and	CCONJ
cana-5407	66	20	the	the	DET
cana-5407	66	21	encoderdecoder	encoderdecoder	NOUN
cana-5407	66	22	with	with	ADP
cana-5407	66	23	a	a	DET
cana-5407	66	24	hybrid	hybrid	ADJ
cana-5407	66	25	connectionist	connectionist	NOUN
cana-5407	66	26	-	-	PUNCT
cana-5407	66	27	temporal	temporal	ADJ
cana-5407	66	28	-	-	PUNCT
cana-5407	66	29	classification	classification	NOUN
cana-5407	66	30	-	-	PUNCT
cana-5407	66	31	attention	attention	NOUN
cana-5407	66	32	ctc	ctc	NOUN
cana-5407	66	33	-	-	PUNCT
cana-5407	66	34	attention	attention	NOUN
cana-5407	66	35	model	model	NOUN
cana-5407	66	36	.	.	PUNCT
cana-5407	67	1	an	an	DET
cana-5407	67	2	encoder	encoder	NOUN
cana-5407	67	3	in	in	ADP
cana-5407	67	4	which	which	PRON
cana-5407	67	5	the	the	DET
cana-5407	67	6	middle	middle	ADJ
cana-5407	67	7	layer	layer	NOUN
cana-5407	67	8	is	be	AUX
cana-5407	67	9	fed	feed	VERB
cana-5407	67	10	by	by	ADP
cana-5407	67	11	a	a	DET
cana-5407	67	12	bne	bne	PROPN
cana-5407	67	13	is	be	AUX
cana-5407	67	14	generated	generate	VERB
cana-5407	67	15	as	as	ADP
cana-5407	67	16	a	a	DET
cana-5407	67	17	by	by	ADP
cana-5407	67	18	-	-	PUNCT
cana-5407	67	19	product	product	NOUN
cana-5407	67	20	of	of	ADP
cana-5407	67	21	the	the	DET
cana-5407	67	22	phoneme	phoneme	NOUN
cana-5407	67	23	recognizer	recognizer	NOUN
cana-5407	67	24	training	training	NOUN
cana-5407	67	25	process	process	NOUN
cana-5407	67	26	.	.	PUNCT
cana-5407	68	1	this	this	DET
cana-5407	68	2	bne	bne	PROPN
cana-5407	68	3	is	be	AUX
cana-5407	68	4	used	use	VERB
cana-5407	68	5	with	with	ADP
cana-5407	68	6	spectral	spectral	ADJ
cana-5407	68	7	data	datum	NOUN
cana-5407	68	8	to	to	PART
cana-5407	68	9	produce	produce	VERB
cana-5407	68	10	middle	middle	ADJ
cana-5407	68	11	representations	representation	NOUN
cana-5407	68	12	that	that	PRON
cana-5407	68	13	are	be	AUX
cana-5407	68	14	speaker	speaker	NOUN
cana-5407	68	15	-	-	PUNCT
cana-5407	68	16	independent	independent	ADJ
cana-5407	68	17	,	,	PUNCT
cana-5407	68	18	battle	battle	NOUN
cana-5407	68	19	-	-	PUNCT
cana-5407	68	20	tested	test	VERB
cana-5407	68	21	,	,	PUNCT
cana-5407	68	22	and	and	CCONJ
cana-5407	68	23	comprehensive	comprehensive	ADJ
cana-5407	68	24	for	for	ADP
cana-5407	68	25	spoken	spoken	ADJ
cana-5407	68	26	language	language	NOUN
cana-5407	68	27	.	.	PUNCT
cana-5407	69	1	the	the	DET
cana-5407	69	2	seq2seq	seq2seq	NOUN
cana-5407	69	3	synthesis	synthesis	NOUN
cana-5407	69	4	model	model	NOUN
cana-5407	69	5	is	be	AUX
cana-5407	69	6	trained	train	VERB
cana-5407	69	7	after	after	ADP
cana-5407	69	8	that	that	PRON
cana-5407	69	9	,	,	PUNCT
cana-5407	69	10	using	use	VERB
cana-5407	69	11	multi	multi	ADJ
cana-5407	69	12	-	-	ADJ
cana-5407	69	13	speaker	speaker	NOUN
cana-5407	69	14	location	location	NOUN
cana-5407	69	15	-	-	PUNCT
cana-5407	69	16	relative	relative	ADJ
cana-5407	69	17	attention	attention	NOUN
cana-5407	69	18	to	to	PART
cana-5407	69	19	recreate	recreate	VERB
cana-5407	69	20	spectral	spectral	ADJ
cana-5407	69	21	qualities	quality	NOUN
cana-5407	69	22	from	from	ADP
cana-5407	69	23	the	the	DET
cana-5407	69	24	bottleneck	bottleneck	NOUN
cana-5407	69	25	data	datum	NOUN
cana-5407	69	26	.	.	PUNCT
cana-5407	70	1	meanwhile	meanwhile	ADV
cana-5407	70	2	,	,	PUNCT
cana-5407	70	3	the	the	DET
cana-5407	70	4	speaker	speaker	NOUN
cana-5407	70	5	representation	representation	NOUN
cana-5407	70	6	is	be	AUX
cana-5407	70	7	raised	raise	VERB
cana-5407	70	8	to	to	PART
cana-5407	70	9	control	control	VERB
cana-5407	70	10	the	the	DET
cana-5407	70	11	speaker	speaker	NOUN
cana-5407	70	12	information	information	NOUN
cana-5407	70	13	in	in	ADP
cana-5407	70	14	the	the	DET
cana-5407	70	15	synthetic	synthetic	ADJ
cana-5407	70	16	voice	voice	NOUN
cana-5407	70	17	.	.	PUNCT
cana-5407	71	1	the	the	DET
cana-5407	71	2	seq2seq	seq2seq	PROPN
cana-5407	71	3	model	model	NOUN
cana-5407	71	4	aligns	align	VERB
cana-5407	71	5	extended	extended	ADJ
cana-5407	71	6	sequences	sequence	NOUN
cana-5407	71	7	,	,	PUNCT
cana-5407	71	8	a	a	DET
cana-5407	71	9	problem	problem	NOUN
cana-5407	71	10	confronted	confront	VERB
cana-5407	71	11	by	by	ADP
cana-5407	71	12	most	most	ADJ
cana-5407	71	13	seq2seq	seq2seq	NOUN
cana-5407	71	14	algorithms	algorithm	NOUN
cana-5407	71	15	.	.	PUNCT
cana-5407	72	1	additionally	additionally	ADV
cana-5407	72	2	,	,	PUNCT
cana-5407	72	3	they	they	PRON
cana-5407	72	4	modify	modify	VERB
cana-5407	72	5	the	the	DET
cana-5407	72	6	synthesizer	synthesizer	NOUN
cana-5407	72	7	using	use	VERB
cana-5407	72	8	a	a	DET
cana-5407	72	9	discretized	discretized	ADJ
cana-5407	72	10	mixture	mixture	NOUN
cana-5407	72	11	of	of	ADP
cana-5407	72	12	logistic	logistic	ADJ
cana-5407	72	13	attention	attention	NOUN
cana-5407	72	14	mechanism	mechanism	NOUN
cana-5407	72	15	.	.	PUNCT
cana-5407	73	1	due	due	ADP
cana-5407	73	2	to	to	ADP
cana-5407	73	3	the	the	DET
cana-5407	73	4	length	length	NOUN
cana-5407	73	5	learning	learn	VERB
cana-5407	73	6	procedure	procedure	NOUN
cana-5407	73	7	of	of	ADP
cana-5407	73	8	phoneme	phoneme	NOUN
cana-5407	73	9	recognition	recognition	NOUN
cana-5407	73	10	utilizing	utilize	VERB
cana-5407	73	11	extensive	extensive	ADJ
cana-5407	73	12	speech	speech	NOUN
cana-5407	73	13	recognition	recognition	NOUN
cana-5407	73	14	data	datum	NOUN
cana-5407	73	15	,	,	PUNCT
cana-5407	73	16	they	they	PRON
cana-5407	73	17	can	can	AUX
cana-5407	73	18	achieve	achieve	VERB
cana-5407	73	19	any	any	PRON
cana-5407	73	20	-	-	PUNCT
cana-5407	73	21	to	to	ADP
cana-5407	73	22	-	-	PUNCT
cana-5407	73	23	many	many	ADJ
cana-5407	73	24	voice	voice	NOUN
cana-5407	73	25	conversion	conversion	NOUN
cana-5407	73	26	using	use	VERB
cana-5407	73	27	the	the	DET
cana-5407	73	28	approach	approach	NOUN
cana-5407	73	29	.	.	PUNCT
cana-5407	74	1	wen	wen	PROPN
cana-5407	74	2	-	-	PROPN
cana-5407	74	3	chin	chin	PROPN
cana-5407	74	4	huang	huang	PROPN
cana-5407	74	5	et	et	PROPN
cana-5407	74	6	al	al	PROPN
cana-5407	75	1	[	[	X
cana-5407	75	2	15	15	NUM
cana-5407	75	3	]	]	PUNCT
cana-5407	75	4	proposed	propose	VERB
cana-5407	75	5	that	that	SCONJ
cana-5407	75	6	one	one	PRON
cana-5407	75	7	could	could	AUX
cana-5407	75	8	transfer	transfer	VERB
cana-5407	75	9	information	information	NOUN
cana-5407	75	10	from	from	ADP
cana-5407	75	11	other	other	ADJ
cana-5407	75	12	speech	speech	NOUN
cana-5407	75	13	processing	processing	NOUN
cana-5407	75	14	tasks	task	NOUN
cana-5407	75	15	,	,	PUNCT
cana-5407	75	16	such	such	ADJ
cana-5407	75	17	as	as	ADP
cana-5407	75	18	text	text	NOUN
cana-5407	75	19	-	-	PUNCT
cana-5407	75	20	to	to	ADP
cana-5407	75	21	-	-	PUNCT
cana-5407	75	22	speech	speech	NOUN
cana-5407	75	23	(	(	PUNCT
cana-5407	75	24	tts	tts	PROPN
cana-5407	75	25	)	)	PUNCT
cana-5407	75	26	and	and	CCONJ
cana-5407	75	27	automated	automate	VERB
cana-5407	75	28	speech	speech	NOUN
cana-5407	75	29	recognition	recognition	NOUN
cana-5407	75	30	(	(	PUNCT
cana-5407	75	31	asr	asr	PROPN
cana-5407	75	32	)	)	PUNCT
cana-5407	75	33	,	,	PUNCT
cana-5407	75	34	for	for	ADP
cana-5407	75	35	which	which	PRON
cana-5407	75	36	large	large	ADJ
cana-5407	75	37	-	-	PUNCT
cana-5407	75	38	scale	scale	NOUN
cana-5407	75	39	collections	collection	NOUN
cana-5407	75	40	of	of	ADP
cana-5407	75	41	high	high	ADJ
cana-5407	75	42	-	-	PUNCT
cana-5407	75	43	quality	quality	NOUN
cana-5407	75	44	data	datum	NOUN
cana-5407	75	45	can	can	AUX
cana-5407	75	46	be	be	AUX
cana-5407	75	47	readily	readily	ADV
cana-5407	75	48	obtained	obtain	VERB
cana-5407	75	49	.	.	PUNCT
cana-5407	76	1	the	the	DET
cana-5407	76	2	authors	author	NOUN
cana-5407	76	3	believe	believe	VERB
cana-5407	76	4	that	that	SCONJ
cana-5407	76	5	when	when	SCONJ
cana-5407	76	6	initiated	initiate	VERB
cana-5407	76	7	with	with	ADP
cana-5407	76	8	parameters	parameter	NOUN
cana-5407	76	9	of	of	ADP
cana-5407	76	10	pretrained	pretraine	VERB
cana-5407	76	11	asr	asr	NOUN
cana-5407	76	12	or	or	CCONJ
cana-5407	76	13	tts	tts	PROPN
cana-5407	76	14	models	model	NOUN
cana-5407	76	15	,	,	PUNCT
cana-5407	76	16	vc	vc	PROPN
cana-5407	76	17	models	model	NOUN
cana-5407	76	18	could	could	AUX
cana-5407	76	19	come	come	VERB
cana-5407	76	20	up	up	ADP
cana-5407	76	21	with	with	ADP
cana-5407	76	22	powerful	powerful	ADJ
cana-5407	76	23	latent	latent	NOUN
cana-5407	76	24	representations	representation	NOUN
cana-5407	76	25	that	that	PRON
cana-5407	76	26	can	can	AUX
cana-5407	76	27	convert	convert	VERB
cana-5407	76	28	speech	speech	NOUN
cana-5407	76	29	that	that	PRON
cana-5407	76	30	is	be	AUX
cana-5407	76	31	both	both	PRON
cana-5407	76	32	of	of	ADP
cana-5407	76	33	high	high	ADJ
cana-5407	76	34	quality	quality	NOUN
cana-5407	76	35	and	and	CCONJ
cana-5407	76	36	highly	highly	ADV
cana-5407	76	37	understandable	understandable	ADJ
cana-5407	76	38	at	at	ADP
cana-5407	76	39	the	the	DET
cana-5407	76	40	same	same	ADJ
cana-5407	76	41	time	time	NOUN
cana-5407	76	42	.	.	PUNCT
cana-5407	77	1	evaluated	evaluate	VERB
cana-5407	77	2	and	and	CCONJ
cana-5407	77	3	addressed	address	VERB
cana-5407	77	4	through	through	ADP
cana-5407	77	5	this	this	DET
cana-5407	77	6	technique	technique	NOUN
cana-5407	77	7	,	,	PUNCT
cana-5407	77	8	which	which	PRON
cana-5407	77	9	is	be	AUX
cana-5407	77	10	used	use	VERB
cana-5407	77	11	in	in	ADP
cana-5407	77	12	a	a	DET
cana-5407	77	13	parallel	parallel	NOUN
cana-5407	77	14	,	,	PUNCT
cana-5407	77	15	one	one	NUM
cana-5407	77	16	-	-	PUNCT
cana-5407	77	17	to	to	ADP
cana-5407	77	18	-	-	PUNCT
cana-5407	77	19	one	one	NUM
cana-5407	77	20	framework	framework	NOUN
cana-5407	77	21	.	.	PUNCT
cana-5407	78	1	they	they	PRON
cana-5407	78	2	applied	apply	VERB
cana-5407	78	3	recurrent	recurrent	ADJ
cana-5407	78	4	neural	neural	ADJ
cana-5407	78	5	networks	network	NOUN
cana-5407	78	6	(	(	PUNCT
cana-5407	78	7	rnn	rnn	PROPN
cana-5407	78	8	)	)	PUNCT
cana-5407	78	9	and	and	CCONJ
cana-5407	78	10	transformer	transformer	NOUN
cana-5407	78	11	models	model	NOUN
cana-5407	78	12	.	.	PUNCT
cana-5407	79	1	communications	communication	NOUN
cana-5407	79	2	on	on	ADP
cana-5407	79	3	applied	apply	VERB
cana-5407	79	4	nonlinear	nonlinear	ADJ
cana-5407	79	5	analysis	analysis	NOUN
cana-5407	79	6	issn	issn	NOUN
cana-5407	79	7	:	:	PUNCT
cana-5407	79	8	1074	1074	NUM
cana-5407	79	9	-	-	PUNCT
cana-5407	79	10	133x	133x	NUM
cana-5407	79	11	vol	vol	VERB
cana-5407	79	12	32	32	NUM
cana-5407	79	13	no	no	NOUN
cana-5407	79	14	.	.	PUNCT
cana-5407	80	1	10s	10	NOUN
cana-5407	80	2	(	(	PUNCT
cana-5407	80	3	2025	2025	NUM
cana-5407	80	4	)	)	PUNCT
cana-5407	80	5	2175	2175	NUM
cana-5407	81	1	https://internationalpubls.com	https://internationalpubls.com	X
cana-5407	81	2	kun	kun	NOUN
cana-5407	81	3	zhou	zhou	PROPN
cana-5407	81	4	et	et	PROPN
cana-5407	81	5	al	al	PROPN
cana-5407	82	1	[	[	X
cana-5407	82	2	16	16	NUM
cana-5407	82	3	]	]	PUNCT
cana-5407	82	4	proposed	propose	VERB
cana-5407	82	5	method	method	NOUN
cana-5407	82	6	attempts	attempt	VERB
cana-5407	82	7	to	to	PART
cana-5407	82	8	seperate	seperate	VERB
cana-5407	82	9	the	the	DET
cana-5407	82	10	manner	manner	NOUN
cana-5407	82	11	of	of	ADP
cana-5407	82	12	speaking	speak	VERB
cana-5407	82	13	from	from	ADP
cana-5407	82	14	the	the	DET
cana-5407	82	15	linguistic	linguistic	ADJ
cana-5407	82	16	content	content	NOUN
cana-5407	82	17	and	and	CCONJ
cana-5407	82	18	model	model	VERB
cana-5407	82	19	the	the	DET
cana-5407	82	20	speaker	speaker	NOUN
cana-5407	82	21	’s	’s	PART
cana-5407	82	22	style	style	NOUN
cana-5407	82	23	as	as	ADP
cana-5407	82	24	a	a	DET
cana-5407	82	25	style	style	NOUN
cana-5407	82	26	embedding	embed	VERB
cana-5407	82	27	in	in	ADP
cana-5407	82	28	a	a	DET
cana-5407	82	29	continuous	continuous	ADJ
cana-5407	82	30	space	space	NOUN
cana-5407	82	31	that	that	PRON
cana-5407	82	32	acts	act	VERB
cana-5407	82	33	as	as	ADP
cana-5407	82	34	the	the	DET
cana-5407	82	35	prototype	prototype	NOUN
cana-5407	82	36	emotion	emotion	NOUN
cana-5407	82	37	embedding	embed	VERB
cana-5407	82	38	.	.	PUNCT
cana-5407	83	1	the	the	DET
cana-5407	83	2	authors	author	NOUN
cana-5407	83	3	obtained	obtain	VERB
cana-5407	83	4	the	the	DET
cana-5407	83	5	accurate	accurate	ADJ
cana-5407	83	6	emotion	emotion	NOUN
cana-5407	83	7	encoder	encoder	NOUN
cana-5407	83	8	with	with	ADP
cana-5407	83	9	the	the	DET
cana-5407	83	10	emotion	emotion	NOUN
cana-5407	83	11	labels	label	NOUN
cana-5407	83	12	from	from	ADP
cana-5407	83	13	the	the	DET
cana-5407	83	14	emotions	emotion	NOUN
cana-5407	83	15	database	database	NOUN
cana-5407	83	16	.	.	PUNCT
cana-5407	84	1	furthermore	furthermore	ADV
cana-5407	84	2	,	,	PUNCT
cana-5407	84	3	they	they	PRON
cana-5407	84	4	also	also	ADV
cana-5407	84	5	analyzed	analyze	VERB
cana-5407	84	6	the	the	DET
cana-5407	84	7	use	use	NOUN
cana-5407	84	8	of	of	ADP
cana-5407	84	9	relative	relative	ADJ
cana-5407	84	10	attributes	attribute	NOUN
cana-5407	84	11	to	to	PART
cana-5407	84	12	name	name	VERB
cana-5407	84	13	the	the	DET
cana-5407	84	14	emotions	emotion	NOUN
cana-5407	84	15	with	with	ADP
cana-5407	84	16	specific	specific	ADJ
cana-5407	84	17	intensity	intensity	NOUN
cana-5407	84	18	.	.	PUNCT
cana-5407	85	1	to	to	PART
cana-5407	85	2	improve	improve	VERB
cana-5407	85	3	personalization	personalization	NOUN
cana-5407	85	4	awareness	awareness	NOUN
cana-5407	85	5	for	for	ADP
cana-5407	85	6	emotions	emotion	NOUN
cana-5407	85	7	,	,	PUNCT
cana-5407	85	8	emotion	emotion	NOUN
cana-5407	85	9	classification	classification	NOUN
cana-5407	85	10	loss	loss	NOUN
cana-5407	85	11	and	and	CCONJ
cana-5407	85	12	emotion	emotion	NOUN
cana-5407	85	13	embedding	embed	VERB
cana-5407	85	14	similarity	similarity	NOUN
cana-5407	85	15	loss	loss	NOUN
cana-5407	85	16	are	be	AUX
cana-5407	85	17	introduced	introduce	VERB
cana-5407	85	18	to	to	ADP
cana-5407	85	19	the	the	DET
cana-5407	85	20	training	training	NOUN
cana-5407	85	21	of	of	ADP
cana-5407	85	22	the	the	DET
cana-5407	85	23	evc	evc	PROPN
cana-5407	85	24	network	network	NOUN
cana-5407	85	25	.	.	PUNCT
cana-5407	86	1	the	the	DET
cana-5407	86	2	network	network	NOUN
cana-5407	86	3	suggested	suggest	VERB
cana-5407	86	4	adequately	adequately	ADV
cana-5407	86	5	controls	control	VERB
cana-5407	86	6	the	the	DET
cana-5407	86	7	correct	correct	ADJ
cana-5407	86	8	emotional	emotional	ADJ
cana-5407	86	9	intensity	intensity	NOUN
cana-5407	86	10	level	level	NOUN
cana-5407	86	11	in	in	ADP
cana-5407	86	12	the	the	DET
cana-5407	86	13	produced	produce	VERB
cana-5407	86	14	speech	speech	NOUN
cana-5407	86	15	.	.	PUNCT
cana-5407	87	1	proposed	propose	VERB
cana-5407	87	2	method	method	NOUN
cana-5407	87	3	:	:	PUNCT
cana-5407	87	4	in	in	ADP
cana-5407	87	5	this	this	DET
cana-5407	87	6	method	method	NOUN
cana-5407	87	7	,	,	PUNCT
cana-5407	87	8	the	the	DET
cana-5407	87	9	initial	initial	ADJ
cana-5407	87	10	audio	audio	ADJ
cana-5407	87	11	input	input	NOUN
cana-5407	87	12	is	be	AUX
cana-5407	87	13	first	first	ADV
cana-5407	87	14	given	give	VERB
cana-5407	87	15	to	to	ADP
cana-5407	87	16	an	an	DET
cana-5407	87	17	automatic	automatic	ADJ
cana-5407	87	18	speech	speech	NOUN
cana-5407	87	19	recognition	recognition	NOUN
cana-5407	87	20	(	(	PUNCT
cana-5407	87	21	asr	asr	NOUN
cana-5407	87	22	)	)	PUNCT
cana-5407	87	23	system	system	NOUN
cana-5407	87	24	,	,	PUNCT
cana-5407	87	25	to	to	PART
cana-5407	87	26	transcribe	transcribe	VERB
cana-5407	87	27	the	the	DET
cana-5407	87	28	spoken	spoken	ADJ
cana-5407	87	29	content	content	NOUN
cana-5407	87	30	into	into	ADP
cana-5407	87	31	text	text	NOUN
cana-5407	87	32	.	.	PUNCT
cana-5407	88	1	subsequently	subsequently	ADV
cana-5407	88	2	,	,	PUNCT
cana-5407	88	3	this	this	DET
cana-5407	88	4	text	text	NOUN
cana-5407	88	5	is	be	AUX
cana-5407	88	6	fed	feed	VERB
cana-5407	88	7	into	into	ADP
cana-5407	88	8	the	the	DET
cana-5407	88	9	voice	voice	NOUN
cana-5407	88	10	conversion	conversion	NOUN
cana-5407	88	11	system	system	NOUN
cana-5407	88	12	.	.	PUNCT
cana-5407	89	1	the	the	DET
cana-5407	89	2	speech	speech	NOUN
cana-5407	89	3	-	-	PUNCT
cana-5407	89	4	to	to	ADP
cana-5407	89	5	-	-	PUNCT
cana-5407	89	6	text	text	NOUN
cana-5407	89	7	component	component	NOUN
cana-5407	89	8	utilizes	utilize	VERB
cana-5407	89	9	python	python	PROPN
cana-5407	89	10	's	's	PART
cana-5407	89	11	speech	speech	NOUN
cana-5407	89	12	recognition	recognition	NOUN
cana-5407	89	13	libraries	library	NOUN
cana-5407	89	14	to	to	PART
cana-5407	89	15	accurately	accurately	ADV
cana-5407	89	16	convert	convert	VERB
cana-5407	89	17	spoken	speak	VERB
cana-5407	89	18	language	language	NOUN
cana-5407	89	19	into	into	ADP
cana-5407	89	20	written	write	VERB
cana-5407	89	21	text	text	NOUN
cana-5407	89	22	.	.	PUNCT
cana-5407	90	1	once	once	SCONJ
cana-5407	90	2	the	the	DET
cana-5407	90	3	transcription	transcription	NOUN
cana-5407	90	4	is	be	AUX
cana-5407	90	5	obtained	obtain	VERB
cana-5407	90	6	,	,	PUNCT
cana-5407	90	7	it	it	PRON
cana-5407	90	8	is	be	AUX
cana-5407	90	9	provided	provide	VERB
cana-5407	90	10	as	as	ADP
cana-5407	90	11	input	input	NOUN
cana-5407	90	12	to	to	ADP
cana-5407	90	13	the	the	DET
cana-5407	90	14	voice	voice	NOUN
cana-5407	90	15	conversion	conversion	PROPN
cana-5407	90	16	model	model	NOUN
cana-5407	90	17	.	.	PUNCT
cana-5407	91	1	the	the	DET
cana-5407	91	2	model	model	NOUN
cana-5407	91	3	then	then	ADV
cana-5407	91	4	embarks	embark	VERB
cana-5407	91	5	on	on	ADP
cana-5407	91	6	its	its	PRON
cana-5407	91	7	synthesis	synthesis	NOUN
cana-5407	91	8	process	process	NOUN
cana-5407	91	9	,	,	PUNCT
cana-5407	91	10	beginning	begin	VERB
cana-5407	91	11	with	with	ADP
cana-5407	91	12	the	the	DET
cana-5407	91	13	phoneme	phoneme	NOUN
cana-5407	91	14	embedding	embed	VERB
cana-5407	91	15	,	,	PUNCT
cana-5407	91	16	followed	follow	VERB
cana-5407	91	17	by	by	ADP
cana-5407	91	18	feature	feature	NOUN
cana-5407	91	19	extraction	extraction	NOUN
cana-5407	91	20	through	through	ADP
cana-5407	91	21	convolutional	convolutional	ADJ
cana-5407	91	22	layers	layer	NOUN
cana-5407	91	23	,	,	PUNCT
cana-5407	91	24	and	and	CCONJ
cana-5407	91	25	sequential	sequential	ADJ
cana-5407	91	26	processing	processing	NOUN
cana-5407	91	27	using	use	VERB
cana-5407	91	28	bidirectional	bidirectional	ADJ
cana-5407	91	29	lstms	lstms	NOUN
cana-5407	91	30	within	within	ADP
cana-5407	91	31	the	the	DET
cana-5407	91	32	encoder	encoder	NOUN
cana-5407	91	33	.	.	PUNCT
cana-5407	92	1	the	the	DET
cana-5407	92	2	decoder	decoder	NOUN
cana-5407	92	3	utilizes	utilize	VERB
cana-5407	92	4	location	location	NOUN
cana-5407	92	5	-	-	PUNCT
cana-5407	92	6	sensitive	sensitive	ADJ
cana-5407	92	7	attention	attention	NOUN
cana-5407	92	8	to	to	PART
cana-5407	92	9	produce	produce	VERB
cana-5407	92	10	a	a	DET
cana-5407	92	11	mel	mel	PROPN
cana-5407	92	12	spectrogram	spectrogram	NOUN
cana-5407	92	13	,	,	PUNCT
cana-5407	92	14	which	which	PRON
cana-5407	92	15	is	be	AUX
cana-5407	92	16	further	far	ADV
cana-5407	92	17	enhanced	enhance	VERB
cana-5407	92	18	by	by	ADP
cana-5407	92	19	post	post	ADJ
cana-5407	92	20	-	-	ADJ
cana-5407	92	21	processing	processing	ADJ
cana-5407	92	22	convolutional	convolutional	ADJ
cana-5407	92	23	layers	layer	NOUN
cana-5407	92	24	.	.	PUNCT
cana-5407	93	1	the	the	DET
cana-5407	93	2	mel	mel	PROPN
cana-5407	93	3	spectrogram	spectrogram	PROPN
cana-5407	93	4	is	be	AUX
cana-5407	93	5	converted	convert	VERB
cana-5407	93	6	into	into	ADP
cana-5407	93	7	an	an	DET
cana-5407	93	8	audible	audible	ADJ
cana-5407	93	9	waveform	waveform	NOUN
cana-5407	93	10	using	use	VERB
cana-5407	93	11	a	a	DET
cana-5407	93	12	wavenet	wavenet	ADJ
cana-5407	93	13	vocoder	vocoder	NOUN
cana-5407	93	14	.	.	PUNCT
cana-5407	94	1	this	this	PRON
cana-5407	94	2	represents	represent	VERB
cana-5407	94	3	the	the	DET
cana-5407	94	4	synthesized	synthesize	VERB
cana-5407	94	5	speech	speech	NOUN
cana-5407	94	6	output	output	NOUN
cana-5407	94	7	,	,	PUNCT
cana-5407	94	8	which	which	PRON
cana-5407	94	9	is	be	AUX
cana-5407	94	10	identical	identical	ADJ
cana-5407	94	11	to	to	ADP
cana-5407	94	12	the	the	DET
cana-5407	94	13	original	original	ADJ
cana-5407	94	14	audio	audio	ADJ
cana-5407	94	15	input	input	NOUN
cana-5407	94	16	in	in	ADP
cana-5407	94	17	form	form	NOUN
cana-5407	94	18	of	of	ADP
cana-5407	94	19	text	text	NOUN
cana-5407	94	20	.	.	PUNCT
cana-5407	95	1	figure	figure	NOUN
cana-5407	95	2	1	1	NUM
cana-5407	95	3	:	:	PUNCT
cana-5407	95	4	proposed	propose	VERB
cana-5407	95	5	framework	framework	NOUN
cana-5407	95	6	communications	communication	NOUN
cana-5407	95	7	on	on	ADP
cana-5407	95	8	applied	apply	VERB
cana-5407	95	9	nonlinear	nonlinear	ADJ
cana-5407	95	10	analysis	analysis	NOUN
cana-5407	95	11	issn	issn	NOUN
cana-5407	95	12	:	:	PUNCT
cana-5407	95	13	1074	1074	NUM
cana-5407	95	14	-	-	PUNCT
cana-5407	95	15	133x	133x	NUM
cana-5407	95	16	vol	vol	VERB
cana-5407	95	17	32	32	NUM
cana-5407	95	18	no	no	NOUN
cana-5407	95	19	.	.	PUNCT
cana-5407	96	1	10s	10	NOUN
cana-5407	96	2	(	(	PUNCT
cana-5407	96	3	2025	2025	NUM
cana-5407	96	4	)	)	PUNCT
cana-5407	96	5	2176	2176	NUM
cana-5407	96	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-5407	96	7	voice	voice	NOUN
cana-5407	96	8	conversion	conversion	NOUN
cana-5407	96	9	module	module	NOUN
cana-5407	96	10	:	:	PUNCT
cana-5407	96	11	the	the	DET
cana-5407	96	12	model	model	NOUN
cana-5407	96	13	first	first	ADV
cana-5407	96	14	processes	process	VERB
cana-5407	96	15	textual	textual	ADJ
cana-5407	96	16	input	input	NOUN
cana-5407	96	17	and	and	CCONJ
cana-5407	96	18	then	then	ADV
cana-5407	96	19	converts	convert	VERB
cana-5407	96	20	it	it	PRON
cana-5407	96	21	into	into	ADP
cana-5407	96	22	a	a	DET
cana-5407	96	23	series	series	NOUN
cana-5407	96	24	of	of	ADP
cana-5407	96	25	phonemes	phoneme	NOUN
cana-5407	96	26	,	,	PUNCT
cana-5407	96	27	which	which	PRON
cana-5407	96	28	are	be	AUX
cana-5407	96	29	translated	translate	VERB
cana-5407	96	30	into	into	ADP
cana-5407	96	31	numerical	numerical	ADJ
cana-5407	96	32	representations	representation	NOUN
cana-5407	96	33	.	.	PUNCT
cana-5407	97	1	the	the	DET
cana-5407	97	2	encoder	encoder	NOUN
cana-5407	97	3	extracts	extract	VERB
cana-5407	97	4	feature	feature	NOUN
cana-5407	97	5	from	from	ADP
cana-5407	97	6	the	the	DET
cana-5407	97	7	representations	representation	NOUN
cana-5407	97	8	.	.	PUNCT
cana-5407	98	1	then	then	ADV
cana-5407	98	2	,	,	PUNCT
cana-5407	98	3	the	the	DET
cana-5407	98	4	decoder	decoder	NOUN
cana-5407	98	5	uses	use	VERB
cana-5407	98	6	the	the	DET
cana-5407	98	7	attention	attention	NOUN
cana-5407	98	8	process	process	NOUN
cana-5407	98	9	to	to	PART
cana-5407	98	10	generate	generate	VERB
cana-5407	98	11	a	a	DET
cana-5407	98	12	mel	mel	PROPN
cana-5407	98	13	spectrogram	spectrogram	NOUN
cana-5407	98	14	.	.	PUNCT
cana-5407	99	1	a	a	DET
cana-5407	99	2	vocoder	vocoder	NOUN
cana-5407	99	3	turns	turn	VERB
cana-5407	99	4	the	the	DET
cana-5407	99	5	mel	mel	PROPN
cana-5407	99	6	spectrogram	spectrogram	NOUN
cana-5407	99	7	into	into	ADP
cana-5407	99	8	a	a	DET
cana-5407	99	9	waveform	waveform	NOUN
cana-5407	99	10	,	,	PUNCT
cana-5407	99	11	which	which	PRON
cana-5407	99	12	can	can	AUX
cana-5407	99	13	be	be	AUX
cana-5407	99	14	spoken	speak	VERB
cana-5407	99	15	as	as	ADP
cana-5407	99	16	the	the	DET
cana-5407	99	17	original	original	ADJ
cana-5407	99	18	input	input	NOUN
cana-5407	99	19	text	text	NOUN
cana-5407	99	20	.	.	PUNCT
cana-5407	100	1	pre	pre	ADJ
cana-5407	100	2	-	-	ADJ
cana-5407	100	3	processing	processing	NOUN
cana-5407	100	4	is	be	AUX
cana-5407	100	5	a	a	DET
cana-5407	100	6	critical	critical	ADJ
cana-5407	100	7	part	part	NOUN
cana-5407	100	8	of	of	ADP
cana-5407	100	9	voice	voice	NOUN
cana-5407	100	10	conversion	conversion	NOUN
cana-5407	100	11	systems	system	NOUN
cana-5407	100	12	because	because	SCONJ
cana-5407	100	13	it	it	PRON
cana-5407	100	14	converts	convert	VERB
cana-5407	100	15	the	the	DET
cana-5407	100	16	raw	raw	ADJ
cana-5407	100	17	input	input	NOUN
cana-5407	100	18	into	into	ADP
cana-5407	100	19	a	a	DET
cana-5407	100	20	form	form	NOUN
cana-5407	100	21	that	that	PRON
cana-5407	100	22	can	can	AUX
cana-5407	100	23	be	be	AUX
cana-5407	100	24	used	use	VERB
cana-5407	100	25	to	to	PART
cana-5407	100	26	train	train	NOUN
cana-5407	100	27	models	model	NOUN
cana-5407	100	28	.	.	PUNCT
cana-5407	101	1	this	this	PRON
cana-5407	101	2	involves	involve	VERB
cana-5407	101	3	converting	convert	VERB
cana-5407	101	4	the	the	DET
cana-5407	101	5	audio	audio	ADJ
cana-5407	101	6	recordings	recording	NOUN
cana-5407	101	7	into	into	ADP
cana-5407	101	8	mel	mel	PROPN
cana-5407	101	9	spectrograms	spectrograms	PROPN
cana-5407	101	10	,	,	PUNCT
cana-5407	101	11	which	which	PRON
cana-5407	101	12	are	be	AUX
cana-5407	101	13	graphical	graphical	ADJ
cana-5407	101	14	representations	representation	NOUN
cana-5407	101	15	of	of	ADP
cana-5407	101	16	the	the	DET
cana-5407	101	17	audio	audio	NOUN
cana-5407	101	18	’s	’s	PART
cana-5407	101	19	fundamental	fundamental	ADJ
cana-5407	101	20	frequencies	frequency	NOUN
cana-5407	101	21	presented	present	VERB
cana-5407	101	22	over	over	ADP
cana-5407	101	23	time	time	NOUN
cana-5407	101	24	.	.	PUNCT
cana-5407	102	1	the	the	DET
cana-5407	102	2	spectrograms	spectrogram	NOUN
cana-5407	102	3	are	be	AUX
cana-5407	102	4	fundamental	fundamental	ADJ
cana-5407	102	5	to	to	ADP
cana-5407	102	6	learning	learn	VERB
cana-5407	102	7	from	from	ADP
cana-5407	102	8	the	the	DET
cana-5407	102	9	model	model	NOUN
cana-5407	102	10	.	.	PUNCT
cana-5407	103	1	the	the	DET
cana-5407	103	2	model	model	NOUN
cana-5407	103	3	proposed	propose	VERB
cana-5407	103	4	includes	include	VERB
cana-5407	103	5	components	component	NOUN
cana-5407	103	6	that	that	PRON
cana-5407	103	7	enable	enable	VERB
cana-5407	103	8	the	the	DET
cana-5407	103	9	automation	automation	NOUN
cana-5407	103	10	of	of	ADP
cana-5407	103	11	the	the	DET
cana-5407	103	12	conversion	conversion	NOUN
cana-5407	103	13	process	process	NOUN
cana-5407	103	14	and	and	CCONJ
cana-5407	103	15	editing	editing	NOUN
cana-5407	103	16	of	of	ADP
cana-5407	103	17	the	the	DET
cana-5407	103	18	file	file	NOUN
cana-5407	103	19	lists	list	NOUN
cana-5407	103	20	that	that	PRON
cana-5407	103	21	define	define	VERB
cana-5407	103	22	the	the	DET
cana-5407	103	23	data	datum	NOUN
cana-5407	103	24	to	to	PART
cana-5407	103	25	use	use	VERB
cana-5407	103	26	for	for	ADP
cana-5407	103	27	training	training	NOUN
cana-5407	103	28	.	.	PUNCT
cana-5407	104	1	ensuring	ensure	VERB
cana-5407	104	2	that	that	SCONJ
cana-5407	104	3	the	the	DET
cana-5407	104	4	data	data	NOUN
cana-5407	104	5	is	be	AUX
cana-5407	104	6	in	in	ADP
cana-5407	104	7	the	the	DET
cana-5407	104	8	right	right	ADJ
cana-5407	104	9	format	format	NOUN
cana-5407	104	10	and	and	CCONJ
cana-5407	104	11	readily	readily	ADV
cana-5407	104	12	accessible	accessible	ADJ
cana-5407	104	13	to	to	ADP
cana-5407	104	14	the	the	DET
cana-5407	104	15	model	model	NOUN
cana-5407	104	16	is	be	AUX
cana-5407	104	17	a	a	DET
cana-5407	104	18	critical	critical	ADJ
cana-5407	104	19	step	step	NOUN
cana-5407	104	20	that	that	PRON
cana-5407	104	21	influences	influence	VERB
cana-5407	104	22	the	the	DET
cana-5407	104	23	training	training	NOUN
cana-5407	104	24	process	process	NOUN
cana-5407	104	25	’s	’s	PART
cana-5407	104	26	efficiency	efficiency	NOUN
cana-5407	104	27	and	and	CCONJ
cana-5407	104	28	effectiveness	effectiveness	NOUN
cana-5407	104	29	directly	directly	ADV
cana-5407	104	30	.	.	PUNCT
cana-5407	105	1	encoder	encoder	NOUN
cana-5407	105	2	phone	phone	NOUN
cana-5407	105	3	embedding	embed	VERB
cana-5407	106	1	:	:	PUNCT
cana-5407	106	2	the	the	DET
cana-5407	106	3	text	text	NOUN
cana-5407	106	4	is	be	AUX
cana-5407	106	5	converted	convert	VERB
cana-5407	106	6	into	into	ADP
cana-5407	106	7	a	a	DET
cana-5407	106	8	phonetic	phonetic	ADJ
cana-5407	106	9	representation	representation	NOUN
cana-5407	106	10	,	,	PUNCT
cana-5407	106	11	where	where	SCONJ
cana-5407	106	12	each	each	DET
cana-5407	106	13	unit	unit	NOUN
cana-5407	106	14	of	of	ADP
cana-5407	106	15	sound	sound	NOUN
cana-5407	106	16	(	(	PUNCT
cana-5407	106	17	phoneme	phoneme	NOUN
cana-5407	106	18	)	)	PUNCT
cana-5407	106	19	is	be	AUX
cana-5407	106	20	transformed	transform	VERB
cana-5407	106	21	into	into	ADP
cana-5407	106	22	a	a	DET
cana-5407	106	23	numerical	numerical	ADJ
cana-5407	106	24	vector	vector	NOUN
cana-5407	106	25	using	use	VERB
cana-5407	106	26	an	an	DET
cana-5407	106	27	embedding	embed	VERB
cana-5407	106	28	layer	layer	NOUN
cana-5407	106	29	.	.	PUNCT
cana-5407	107	1	this	this	PRON
cana-5407	107	2	helps	help	VERB
cana-5407	107	3	the	the	DET
cana-5407	107	4	model	model	NOUN
cana-5407	107	5	understand	understand	VERB
cana-5407	107	6	the	the	DET
cana-5407	107	7	text	text	NOUN
cana-5407	107	8	in	in	ADP
cana-5407	107	9	terms	term	NOUN
cana-5407	107	10	of	of	ADP
cana-5407	107	11	sounds	sound	NOUN
cana-5407	107	12	rather	rather	ADV
cana-5407	107	13	than	than	ADP
cana-5407	107	14	letters	letter	NOUN
cana-5407	107	15	.	.	PUNCT
cana-5407	108	1	convolution	convolution	NOUN
cana-5407	108	2	layers	layer	NOUN
cana-5407	108	3	:	:	PUNCT
cana-5407	108	4	these	these	DET
cana-5407	108	5	layers	layer	NOUN
cana-5407	108	6	are	be	AUX
cana-5407	108	7	used	use	VERB
cana-5407	108	8	to	to	PART
cana-5407	108	9	extract	extract	VERB
cana-5407	108	10	features	feature	NOUN
cana-5407	108	11	from	from	ADP
cana-5407	108	12	the	the	DET
cana-5407	108	13	phoneme	phoneme	NOUN
cana-5407	108	14	embeddings	embedding	NOUN
cana-5407	108	15	.	.	PUNCT
cana-5407	109	1	convolutions	convolution	NOUN
cana-5407	109	2	help	help	VERB
cana-5407	109	3	the	the	DET
cana-5407	109	4	model	model	NOUN
cana-5407	109	5	to	to	PART
cana-5407	109	6	capture	capture	VERB
cana-5407	109	7	the	the	DET
cana-5407	109	8	local	local	ADJ
cana-5407	109	9	context	context	NOUN
cana-5407	109	10	within	within	ADP
cana-5407	109	11	the	the	DET
cana-5407	109	12	sequence	sequence	NOUN
cana-5407	109	13	of	of	ADP
cana-5407	109	14	phonemes	phoneme	NOUN
cana-5407	109	15	.	.	PUNCT
cana-5407	110	1	bidirectional	bidirectional	ADJ
cana-5407	110	2	lstm	lstm	NOUN
cana-5407	110	3	:	:	PUNCT
cana-5407	110	4	a	a	DET
cana-5407	110	5	type	type	NOUN
cana-5407	110	6	of	of	ADP
cana-5407	110	7	recurrent	recurrent	ADJ
cana-5407	110	8	neural	neural	ADJ
cana-5407	110	9	network	network	NOUN
cana-5407	110	10	that	that	PRON
cana-5407	110	11	processes	process	VERB
cana-5407	110	12	the	the	DET
cana-5407	110	13	sequence	sequence	NOUN
cana-5407	110	14	of	of	ADP
cana-5407	110	15	features	feature	NOUN
cana-5407	110	16	from	from	ADP
cana-5407	110	17	both	both	DET
cana-5407	110	18	directions	direction	NOUN
cana-5407	110	19	(	(	PUNCT
cana-5407	110	20	forward	forward	ADV
cana-5407	110	21	and	and	CCONJ
cana-5407	110	22	backward	backward	ADJ
cana-5407	110	23	)	)	PUNCT
cana-5407	110	24	.	.	PUNCT
cana-5407	111	1	this	this	PRON
cana-5407	111	2	ensures	ensure	VERB
cana-5407	111	3	that	that	SCONJ
cana-5407	111	4	the	the	DET
cana-5407	111	5	model	model	NOUN
cana-5407	111	6	has	have	VERB
cana-5407	111	7	access	access	NOUN
cana-5407	111	8	to	to	ADP
cana-5407	111	9	past	past	NOUN
cana-5407	111	10	and	and	CCONJ
cana-5407	111	11	future	future	ADJ
cana-5407	111	12	context	context	NOUN
cana-5407	111	13	,	,	PUNCT
cana-5407	111	14	which	which	PRON
cana-5407	111	15	is	be	AUX
cana-5407	111	16	crucial	crucial	ADJ
cana-5407	111	17	for	for	ADP
cana-5407	111	18	predicting	predict	VERB
cana-5407	111	19	the	the	DET
cana-5407	111	20	correct	correct	ADJ
cana-5407	111	21	intonation	intonation	NOUN
cana-5407	111	22	and	and	CCONJ
cana-5407	111	23	rhythm	rhythm	NOUN
cana-5407	111	24	of	of	ADP
cana-5407	111	25	speech	speech	NOUN
cana-5407	111	26	.	.	PUNCT
cana-5407	112	1	decoder	decoder	NOUN
cana-5407	112	2	:	:	PUNCT
cana-5407	112	3	the	the	DET
cana-5407	112	4	decoder	decoder	NOUN
cana-5407	112	5	is	be	AUX
cana-5407	112	6	responsible	responsible	ADJ
cana-5407	112	7	for	for	ADP
cana-5407	112	8	generating	generate	VERB
cana-5407	112	9	the	the	DET
cana-5407	112	10	mel	mel	PROPN
cana-5407	112	11	spectrogram	spectrogram	PROPN
cana-5407	112	12	from	from	ADP
cana-5407	112	13	the	the	DET
cana-5407	112	14	encoded	encode	VERB
cana-5407	112	15	features	feature	NOUN
cana-5407	112	16	.	.	PUNCT
cana-5407	113	1	two	two	NUM
cana-5407	113	2	layer	layer	NOUN
cana-5407	113	3	pre	pre	NOUN
cana-5407	113	4	-	-	NOUN
cana-5407	113	5	net	net	ADJ
cana-5407	113	6	:	:	PUNCT
cana-5407	113	7	before	before	ADP
cana-5407	113	8	processing	process	VERB
cana-5407	113	9	in	in	ADP
cana-5407	113	10	the	the	DET
cana-5407	113	11	lstm	lstm	ADJ
cana-5407	113	12	layers	layer	NOUN
cana-5407	113	13	,	,	PUNCT
cana-5407	113	14	the	the	DET
cana-5407	113	15	inputs	input	NOUN
cana-5407	113	16	pass	pass	VERB
cana-5407	113	17	through	through	ADP
cana-5407	113	18	a	a	DET
cana-5407	113	19	small	small	ADJ
cana-5407	113	20	feedforward	feedforward	NOUN
cana-5407	113	21	neural	neural	ADJ
cana-5407	113	22	network	network	NOUN
cana-5407	113	23	called	call	VERB
cana-5407	113	24	a	a	DET
cana-5407	113	25	pre	pre	NOUN
cana-5407	113	26	-	-	NOUN
cana-5407	113	27	net	net	ADJ
cana-5407	113	28	,	,	PUNCT
cana-5407	113	29	which	which	PRON
cana-5407	113	30	helps	help	VERB
cana-5407	113	31	to	to	PART
cana-5407	113	32	transform	transform	VERB
cana-5407	113	33	the	the	DET
cana-5407	113	34	inputs	input	NOUN
cana-5407	113	35	into	into	ADP
cana-5407	113	36	a	a	DET
cana-5407	113	37	more	more	ADV
cana-5407	113	38	useful	useful	ADJ
cana-5407	113	39	representation	representation	NOUN
cana-5407	113	40	for	for	ADP
cana-5407	113	41	the	the	DET
cana-5407	113	42	lstm	lstm	ADJ
cana-5407	113	43	layers	layer	NOUN
cana-5407	113	44	.	.	PUNCT
cana-5407	114	1	two	two	NUM
cana-5407	114	2	lstm	lstm	ADJ
cana-5407	114	3	layers	layer	NOUN
cana-5407	114	4	:	:	PUNCT
cana-5407	114	5	these	these	PRON
cana-5407	114	6	are	be	AUX
cana-5407	114	7	the	the	DET
cana-5407	114	8	main	main	ADJ
cana-5407	114	9	recurrent	recurrent	ADJ
cana-5407	114	10	layers	layer	NOUN
cana-5407	114	11	of	of	ADP
cana-5407	114	12	the	the	DET
cana-5407	114	13	decoder	decoder	NOUN
cana-5407	114	14	.	.	PUNCT
cana-5407	115	1	they	they	PRON
cana-5407	115	2	process	process	VERB
cana-5407	115	3	the	the	DET
cana-5407	115	4	sequence	sequence	NOUN
cana-5407	115	5	step	step	NOUN
cana-5407	115	6	-	-	PUNCT
cana-5407	115	7	by	by	ADP
cana-5407	115	8	-	-	PUNCT
cana-5407	115	9	step	step	NOUN
cana-5407	115	10	and	and	CCONJ
cana-5407	115	11	output	output	VERB
cana-5407	115	12	a	a	DET
cana-5407	115	13	new	new	ADJ
cana-5407	115	14	representation	representation	NOUN
cana-5407	115	15	that	that	PRON
cana-5407	115	16	captures	capture	VERB
cana-5407	115	17	the	the	DET
cana-5407	115	18	temporal	temporal	ADJ
cana-5407	115	19	structure	structure	NOUN
cana-5407	115	20	of	of	ADP
cana-5407	115	21	speech	speech	NOUN
cana-5407	115	22	.	.	PUNCT
cana-5407	116	1	location	location	NOUN
cana-5407	116	2	sensitive	sensitive	ADJ
cana-5407	116	3	attention	attention	NOUN
cana-5407	116	4	:	:	PUNCT
cana-5407	116	5	this	this	DET
cana-5407	116	6	component	component	NOUN
cana-5407	116	7	helps	help	VERB
cana-5407	116	8	the	the	DET
cana-5407	116	9	decoder	decoder	NOUN
cana-5407	116	10	focus	focus	VERB
cana-5407	116	11	on	on	ADP
cana-5407	116	12	different	different	ADJ
cana-5407	116	13	parts	part	NOUN
cana-5407	116	14	of	of	ADP
cana-5407	116	15	the	the	DET
cana-5407	116	16	encoder	encoder	NOUN
cana-5407	116	17	output	output	NOUN
cana-5407	116	18	at	at	ADP
cana-5407	116	19	each	each	DET
cana-5407	116	20	step	step	NOUN
cana-5407	116	21	of	of	ADP
cana-5407	116	22	the	the	DET
cana-5407	116	23	generation	generation	NOUN
cana-5407	116	24	process	process	NOUN
cana-5407	116	25	.	.	PUNCT
cana-5407	117	1	it	it	PRON
cana-5407	117	2	's	be	AUX
cana-5407	117	3	"	"	PUNCT
cana-5407	117	4	location	location	NOUN
cana-5407	117	5	-	-	PUNCT
cana-5407	117	6	sensitive	sensitive	ADJ
cana-5407	117	7	"	"	PUNCT
cana-5407	117	8	because	because	SCONJ
cana-5407	117	9	it	it	PRON
cana-5407	117	10	takes	take	VERB
cana-5407	117	11	into	into	ADP
cana-5407	117	12	account	account	NOUN
cana-5407	117	13	the	the	DET
cana-5407	117	14	decoder	decoder	NOUN
cana-5407	117	15	's	's	PART
cana-5407	117	16	previous	previous	ADJ
cana-5407	117	17	focus	focus	NOUN
cana-5407	117	18	points	point	NOUN
cana-5407	117	19	to	to	PART
cana-5407	117	20	inform	inform	VERB
cana-5407	117	21	its	its	PRON
cana-5407	117	22	current	current	ADJ
cana-5407	117	23	focus	focus	NOUN
cana-5407	117	24	,	,	PUNCT
cana-5407	117	25	which	which	PRON
cana-5407	117	26	is	be	AUX
cana-5407	117	27	important	important	ADJ
cana-5407	117	28	for	for	ADP
cana-5407	117	29	aligning	align	VERB
cana-5407	117	30	the	the	DET
cana-5407	117	31	generated	generate	VERB
cana-5407	117	32	speech	speech	NOUN
cana-5407	117	33	with	with	ADP
cana-5407	117	34	the	the	DET
cana-5407	117	35	input	input	NOUN
cana-5407	117	36	text	text	NOUN
cana-5407	117	37	.	.	PUNCT
cana-5407	118	1	linear	linear	PROPN
cana-5407	118	2	projection	projection	NOUN
cana-5407	118	3	:	:	PUNCT
cana-5407	118	4	these	these	DET
cana-5407	118	5	layers	layer	NOUN
cana-5407	118	6	project	project	VERB
cana-5407	118	7	the	the	DET
cana-5407	118	8	output	output	NOUN
cana-5407	118	9	of	of	ADP
cana-5407	118	10	the	the	DET
cana-5407	118	11	lstm	lstm	ADJ
cana-5407	118	12	layers	layer	NOUN
cana-5407	118	13	into	into	ADP
cana-5407	118	14	the	the	DET
cana-5407	118	15	mel	mel	PROPN
cana-5407	118	16	spectrogram	spectrogram	NOUN
cana-5407	118	17	space	space	NOUN
cana-5407	118	18	,	,	PUNCT
cana-5407	118	19	creating	create	VERB
cana-5407	118	20	a	a	DET
cana-5407	118	21	raw	raw	ADJ
cana-5407	118	22	mel	mel	PROPN
cana-5407	118	23	spectrogram	spectrogram	NOUN
cana-5407	118	24	prediction	prediction	NOUN
cana-5407	118	25	for	for	ADP
cana-5407	118	26	each	each	DET
cana-5407	118	27	time	time	NOUN
cana-5407	118	28	step	step	NOUN
cana-5407	118	29	.	.	PUNCT
cana-5407	119	1	mel	mel	PROPN
cana-5407	119	2	spectrogram	spectrogram	PROPN
cana-5407	119	3	linear	linear	PROPN
cana-5407	119	4	projection	projection	NOUN
cana-5407	119	5	:	:	PUNCT
cana-5407	119	6	the	the	DET
cana-5407	119	7	first	first	ADJ
cana-5407	119	8	linear	linear	PROPN
cana-5407	119	9	projection	projection	NOUN
cana-5407	119	10	block	block	NOUN
cana-5407	119	11	that	that	PRON
cana-5407	119	12	connects	connect	VERB
cana-5407	119	13	directly	directly	ADV
cana-5407	119	14	to	to	ADP
cana-5407	119	15	the	the	DET
cana-5407	119	16	lstm	lstm	ADJ
cana-5407	119	17	layers	layer	NOUN
cana-5407	119	18	in	in	ADP
cana-5407	119	19	the	the	DET
cana-5407	119	20	decoder	decoder	NOUN
cana-5407	119	21	transforms	transform	VERB
cana-5407	119	22	the	the	DET
cana-5407	119	23	lstm	lstm	ADJ
cana-5407	119	24	output	output	NOUN
cana-5407	119	25	into	into	ADP
cana-5407	119	26	a	a	DET
cana-5407	119	27	predicted	predict	VERB
cana-5407	119	28	mel	mel	PROPN
cana-5407	119	29	spectrogram	spectrogram	PROPN
cana-5407	119	30	.	.	PUNCT
cana-5407	120	1	each	each	DET
cana-5407	120	2	lstm	lstm	ADJ
cana-5407	120	3	output	output	NOUN
cana-5407	120	4	represents	represent	VERB
cana-5407	120	5	a	a	DET
cana-5407	120	6	moment	moment	NOUN
cana-5407	120	7	in	in	ADP
cana-5407	120	8	time	time	NOUN
cana-5407	120	9	within	within	ADP
cana-5407	120	10	the	the	DET
cana-5407	120	11	generated	generate	VERB
cana-5407	120	12	speech	speech	NOUN
cana-5407	120	13	.	.	PUNCT
cana-5407	121	1	this	this	DET
cana-5407	121	2	projection	projection	NOUN
cana-5407	121	3	serves	serve	VERB
cana-5407	121	4	to	to	ADP
cana-5407	121	5	communications	communication	NOUN
cana-5407	121	6	on	on	ADP
cana-5407	121	7	applied	apply	VERB
cana-5407	121	8	nonlinear	nonlinear	ADJ
cana-5407	121	9	analysis	analysis	NOUN
cana-5407	121	10	issn	issn	NOUN
cana-5407	121	11	:	:	PUNCT
cana-5407	121	12	1074	1074	NUM
cana-5407	121	13	-	-	PUNCT
cana-5407	121	14	133x	133x	NUM
cana-5407	121	15	vol	vol	VERB
cana-5407	121	16	32	32	NUM
cana-5407	121	17	no	no	NOUN
cana-5407	121	18	.	.	PUNCT
cana-5407	122	1	10s	10	NOUN
cana-5407	122	2	(	(	PUNCT
cana-5407	122	3	2025	2025	NUM
cana-5407	122	4	)	)	PUNCT
cana-5407	122	5	2177	2177	NUM
cana-5407	122	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-5407	122	7	translate	translate	VERB
cana-5407	122	8	these	these	DET
cana-5407	122	9	outputs	output	NOUN
cana-5407	122	10	into	into	ADP
cana-5407	122	11	a	a	DET
cana-5407	122	12	format	format	NOUN
cana-5407	122	13	that	that	PRON
cana-5407	122	14	represents	represent	VERB
cana-5407	122	15	the	the	DET
cana-5407	122	16	frequency	frequency	NOUN
cana-5407	122	17	content	content	NOUN
cana-5407	122	18	of	of	ADP
cana-5407	122	19	the	the	DET
cana-5407	122	20	audio	audio	ADJ
cana-5407	122	21	signal	signal	NOUN
cana-5407	122	22	at	at	ADP
cana-5407	122	23	each	each	DET
cana-5407	122	24	time	time	NOUN
cana-5407	122	25	step	step	NOUN
cana-5407	122	26	,	,	PUNCT
cana-5407	122	27	which	which	PRON
cana-5407	122	28	is	be	AUX
cana-5407	122	29	what	what	PRON
cana-5407	122	30	a	a	DET
cana-5407	122	31	mel	mel	PROPN
cana-5407	122	32	spectrogram	spectrogram	NOUN
cana-5407	122	33	encapsulates	encapsulate	VERB
cana-5407	122	34	.	.	PUNCT
cana-5407	123	1	stop	stop	VERB
cana-5407	123	2	token	token	VERB
cana-5407	123	3	linear	linear	ADJ
cana-5407	123	4	projection	projection	NOUN
cana-5407	123	5	:	:	PUNCT
cana-5407	123	6	the	the	DET
cana-5407	123	7	second	second	ADJ
cana-5407	123	8	linear	linear	PROPN
cana-5407	123	9	projection	projection	NOUN
cana-5407	123	10	block	block	NOUN
cana-5407	123	11	is	be	AUX
cana-5407	123	12	responsible	responsible	ADJ
cana-5407	123	13	for	for	ADP
cana-5407	123	14	predicting	predict	VERB
cana-5407	123	15	stop	stop	VERB
cana-5407	123	16	tokens	token	NOUN
cana-5407	123	17	.	.	PUNCT
cana-5407	124	1	it	it	PRON
cana-5407	124	2	projects	project	VERB
cana-5407	124	3	the	the	DET
cana-5407	124	4	output	output	NOUN
cana-5407	124	5	of	of	ADP
cana-5407	124	6	the	the	DET
cana-5407	124	7	lstm	lstm	NOUN
cana-5407	124	8	layers	layer	NOUN
cana-5407	124	9	into	into	ADP
cana-5407	124	10	a	a	DET
cana-5407	124	11	binary	binary	ADJ
cana-5407	124	12	decision	decision	NOUN
cana-5407	124	13	for	for	ADP
cana-5407	124	14	each	each	DET
cana-5407	124	15	frame	frame	NOUN
cana-5407	124	16	,	,	PUNCT
cana-5407	124	17	indicating	indicate	VERB
cana-5407	124	18	whether	whether	SCONJ
cana-5407	124	19	the	the	DET
cana-5407	124	20	frame	frame	NOUN
cana-5407	124	21	should	should	AUX
cana-5407	124	22	be	be	AUX
cana-5407	124	23	the	the	DET
cana-5407	124	24	last	last	ADJ
cana-5407	124	25	one	one	NUM
cana-5407	124	26	in	in	ADP
cana-5407	124	27	the	the	DET
cana-5407	124	28	generated	generate	VERB
cana-5407	124	29	mel	mel	PROPN
cana-5407	124	30	spectrogram	spectrogram	PROPN
cana-5407	124	31	(	(	PUNCT
cana-5407	124	32	i.e.	i.e.	X
cana-5407	124	33	,	,	PUNCT
cana-5407	124	34	where	where	SCONJ
cana-5407	124	35	the	the	DET
cana-5407	124	36	speech	speech	NOUN
cana-5407	124	37	should	should	AUX
cana-5407	124	38	stop	stop	VERB
cana-5407	124	39	)	)	PUNCT
cana-5407	124	40	.	.	PUNCT
cana-5407	125	1	this	this	PRON
cana-5407	125	2	is	be	AUX
cana-5407	125	3	crucial	crucial	ADJ
cana-5407	125	4	for	for	SCONJ
cana-5407	125	5	the	the	DET
cana-5407	125	6	model	model	NOUN
cana-5407	125	7	to	to	PART
cana-5407	125	8	be	be	AUX
cana-5407	125	9	able	able	ADJ
cana-5407	125	10	to	to	PART
cana-5407	125	11	generate	generate	VERB
cana-5407	125	12	audio	audio	NOUN
cana-5407	125	13	of	of	ADP
cana-5407	125	14	variable	variable	ADJ
cana-5407	125	15	length	length	NOUN
cana-5407	125	16	that	that	PRON
cana-5407	125	17	corresponds	correspond	VERB
cana-5407	125	18	appropriately	appropriately	ADV
cana-5407	125	19	to	to	ADP
cana-5407	125	20	sentences	sentence	NOUN
cana-5407	125	21	of	of	ADP
cana-5407	125	22	different	different	ADJ
cana-5407	125	23	lengths	length	NOUN
cana-5407	125	24	.	.	PUNCT
cana-5407	126	1	in	in	ADP
cana-5407	126	2	essence	essence	NOUN
cana-5407	126	3	,	,	PUNCT
cana-5407	126	4	while	while	SCONJ
cana-5407	126	5	both	both	DET
cana-5407	126	6	linear	linear	ADJ
cana-5407	126	7	projection	projection	NOUN
cana-5407	126	8	blocks	block	NOUN
cana-5407	126	9	utilize	utilize	VERB
cana-5407	126	10	the	the	DET
cana-5407	126	11	output	output	NOUN
cana-5407	126	12	of	of	ADP
cana-5407	126	13	the	the	DET
cana-5407	126	14	lstm	lstm	ADJ
cana-5407	126	15	layers	layer	NOUN
cana-5407	126	16	,	,	PUNCT
cana-5407	126	17	they	they	PRON
cana-5407	126	18	project	project	VERB
cana-5407	126	19	this	this	DET
cana-5407	126	20	output	output	NOUN
cana-5407	126	21	into	into	ADP
cana-5407	126	22	different	different	ADJ
cana-5407	126	23	spaces	space	NOUN
cana-5407	126	24	for	for	ADP
cana-5407	126	25	different	different	ADJ
cana-5407	126	26	purposes	purpose	NOUN
cana-5407	126	27	:	:	PUNCT
cana-5407	126	28	one	one	NUM
cana-5407	126	29	for	for	ADP
cana-5407	126	30	the	the	DET
cana-5407	126	31	continuous	continuous	ADJ
cana-5407	126	32	mel	mel	PROPN
cana-5407	126	33	spectrogram	spectrogram	PROPN
cana-5407	126	34	frames	frame	NOUN
cana-5407	126	35	,	,	PUNCT
cana-5407	126	36	and	and	CCONJ
cana-5407	126	37	the	the	DET
cana-5407	126	38	other	other	ADJ
cana-5407	126	39	for	for	ADP
cana-5407	126	40	the	the	DET
cana-5407	126	41	binary	binary	PROPN
cana-5407	126	42	stop	stop	PROPN
cana-5407	126	43	tokens	token	NOUN
cana-5407	126	44	.	.	PUNCT
cana-5407	127	1	this	this	PRON
cana-5407	127	2	allows	allow	VERB
cana-5407	127	3	the	the	DET
cana-5407	127	4	decoder	decoder	NOUN
cana-5407	127	5	to	to	PART
cana-5407	127	6	simultaneously	simultaneously	ADV
cana-5407	127	7	predict	predict	VERB
cana-5407	127	8	what	what	PRON
cana-5407	127	9	the	the	DET
cana-5407	127	10	speech	speech	NOUN
cana-5407	127	11	should	should	AUX
cana-5407	127	12	sound	sound	VERB
cana-5407	127	13	like	like	ADP
cana-5407	127	14	at	at	ADP
cana-5407	127	15	each	each	DET
cana-5407	127	16	time	time	NOUN
cana-5407	127	17	step	step	NOUN
cana-5407	127	18	(	(	PUNCT
cana-5407	127	19	the	the	DET
cana-5407	127	20	mel	mel	PROPN
cana-5407	127	21	spectrogram	spectrogram	PROPN
cana-5407	127	22	)	)	PUNCT
cana-5407	127	23	and	and	CCONJ
cana-5407	127	24	when	when	SCONJ
cana-5407	127	25	the	the	DET
cana-5407	127	26	speech	speech	NOUN
cana-5407	127	27	should	should	AUX
cana-5407	127	28	end	end	VERB
cana-5407	127	29	(	(	PUNCT
cana-5407	127	30	the	the	DET
cana-5407	127	31	stop	stop	NOUN
cana-5407	127	32	tokens	token	NOUN
cana-5407	127	33	)	)	PUNCT
cana-5407	127	34	.	.	PUNCT
cana-5407	128	1	conv	conv	ADJ
cana-5407	128	2	layer	layer	NOUN
cana-5407	128	3	post	post	NOUN
cana-5407	128	4	-	-	ADJ
cana-5407	128	5	net	net	ADJ
cana-5407	128	6	:	:	PUNCT
cana-5407	128	7	a	a	DET
cana-5407	128	8	small	small	ADJ
cana-5407	128	9	convolutional	convolutional	ADJ
cana-5407	128	10	network	network	NOUN
cana-5407	128	11	that	that	PRON
cana-5407	128	12	refines	refine	VERB
cana-5407	128	13	the	the	DET
cana-5407	128	14	raw	raw	ADJ
cana-5407	128	15	mel	mel	PROPN
cana-5407	128	16	spectrogram	spectrogram	PROPN
cana-5407	128	17	prediction	prediction	NOUN
cana-5407	128	18	,	,	PUNCT
cana-5407	128	19	improving	improve	VERB
cana-5407	128	20	the	the	DET
cana-5407	128	21	quality	quality	NOUN
cana-5407	128	22	of	of	ADP
cana-5407	128	23	the	the	DET
cana-5407	128	24	audio	audio	NOUN
cana-5407	128	25	that	that	PRON
cana-5407	128	26	will	will	AUX
cana-5407	128	27	be	be	AUX
cana-5407	128	28	synthesized	synthesize	VERB
cana-5407	128	29	.	.	PUNCT
cana-5407	129	1	3.1.3	3.1.3	NUM
cana-5407	129	2	wavenet	wavenet	NOUN
cana-5407	129	3	vocoder	vocoder	NOUN
cana-5407	129	4	the	the	DET
cana-5407	129	5	refined	refined	PROPN
cana-5407	129	6	mel	mel	PROPN
cana-5407	129	7	spectrogram	spectrogram	PROPN
cana-5407	129	8	is	be	AUX
cana-5407	129	9	then	then	ADV
cana-5407	129	10	passed	pass	VERB
cana-5407	129	11	to	to	ADP
cana-5407	129	12	a	a	DET
cana-5407	129	13	wavenet	wavenet	ADJ
cana-5407	129	14	vocoder	vocoder	NOUN
cana-5407	129	15	,	,	PUNCT
cana-5407	129	16	which	which	PRON
cana-5407	129	17	is	be	AUX
cana-5407	129	18	a	a	DET
cana-5407	129	19	neural	neural	ADJ
cana-5407	129	20	network	network	NOUN
cana-5407	129	21	model	model	NOUN
cana-5407	129	22	that	that	PRON
cana-5407	129	23	generates	generate	VERB
cana-5407	129	24	the	the	DET
cana-5407	129	25	final	final	ADJ
cana-5407	129	26	waveform	waveform	NOUN
cana-5407	129	27	(	(	PUNCT
cana-5407	129	28	audio	audio	NOUN
cana-5407	129	29	signal	signal	NOUN
cana-5407	129	30	)	)	PUNCT
cana-5407	129	31	.	.	PUNCT
cana-5407	130	1	the	the	DET
cana-5407	130	2	vocoder	vocoder	NOUN
cana-5407	130	3	translates	translate	VERB
cana-5407	130	4	the	the	DET
cana-5407	130	5	mel	mel	PROPN
cana-5407	130	6	spectrogram	spectrogram	NOUN
cana-5407	130	7	into	into	ADP
cana-5407	130	8	an	an	DET
cana-5407	130	9	audio	audio	ADJ
cana-5407	130	10	waveform	waveform	NOUN
cana-5407	130	11	that	that	PRON
cana-5407	130	12	we	we	PRON
cana-5407	130	13	can	can	AUX
cana-5407	130	14	hear	hear	VERB
cana-5407	130	15	.	.	PUNCT
cana-5407	131	1	3.1.4	3.1.4	NUM
cana-5407	131	2	stop	stop	VERB
cana-5407	131	3	tokens	token	NOUN
cana-5407	131	4	during	during	ADP
cana-5407	131	5	decoding	decode	VERB
cana-5407	131	6	,	,	PUNCT
cana-5407	131	7	the	the	DET
cana-5407	131	8	model	model	NOUN
cana-5407	131	9	also	also	ADV
cana-5407	131	10	predicts	predict	VERB
cana-5407	131	11	stop	stop	VERB
cana-5407	131	12	tokens	token	NOUN
cana-5407	131	13	,	,	PUNCT
cana-5407	131	14	which	which	PRON
cana-5407	131	15	indicate	indicate	VERB
cana-5407	131	16	when	when	SCONJ
cana-5407	131	17	the	the	DET
cana-5407	131	18	generation	generation	NOUN
cana-5407	131	19	of	of	ADP
cana-5407	131	20	the	the	DET
cana-5407	131	21	mel	mel	PROPN
cana-5407	131	22	spectrogram	spectrogram	PROPN
cana-5407	131	23	frames	frame	NOUN
cana-5407	131	24	should	should	AUX
cana-5407	131	25	end	end	VERB
cana-5407	131	26	.	.	PUNCT
cana-5407	132	1	this	this	PRON
cana-5407	132	2	is	be	AUX
cana-5407	132	3	essential	essential	ADJ
cana-5407	132	4	to	to	PART
cana-5407	132	5	determine	determine	VERB
cana-5407	132	6	the	the	DET
cana-5407	132	7	end	end	NOUN
cana-5407	132	8	of	of	ADP
cana-5407	132	9	the	the	DET
cana-5407	132	10	spoken	spoken	ADJ
cana-5407	132	11	output	output	NOUN
cana-5407	132	12	.	.	PUNCT
cana-5407	133	1	3.1.5	3.1.5	NUM
cana-5407	133	2	final	final	ADJ
cana-5407	133	3	output	output	NOUN
cana-5407	133	4	the	the	DET
cana-5407	133	5	final	final	ADJ
cana-5407	133	6	output	output	NOUN
cana-5407	133	7	is	be	AUX
cana-5407	133	8	the	the	DET
cana-5407	133	9	waveform	waveform	NOUN
cana-5407	133	10	samples	sample	NOUN
cana-5407	133	11	,	,	PUNCT
cana-5407	133	12	which	which	PRON
cana-5407	133	13	are	be	AUX
cana-5407	133	14	the	the	DET
cana-5407	133	15	audio	audio	ADJ
cana-5407	133	16	samples	sample	NOUN
cana-5407	133	17	that	that	PRON
cana-5407	133	18	can	can	AUX
cana-5407	133	19	be	be	AUX
cana-5407	133	20	played	play	VERB
cana-5407	133	21	back	back	ADV
cana-5407	133	22	to	to	PART
cana-5407	133	23	produce	produce	VERB
cana-5407	133	24	the	the	DET
cana-5407	133	25	synthesized	synthesize	VERB
cana-5407	133	26	speech	speech	NOUN
cana-5407	133	27	.	.	PUNCT
cana-5407	134	1	3.2	3.2	NUM
cana-5407	134	2	hyperparameter	hyperparameter	NOUN
cana-5407	134	3	settings	setting	NOUN
cana-5407	134	4	hyperparameters	hyperparameter	NOUN
cana-5407	134	5	are	be	AUX
cana-5407	134	6	the	the	DET
cana-5407	134	7	configurable	configurable	ADJ
cana-5407	134	8	variables	variable	NOUN
cana-5407	134	9	used	use	VERB
cana-5407	134	10	to	to	PART
cana-5407	134	11	control	control	VERB
cana-5407	134	12	the	the	DET
cana-5407	134	13	training	training	NOUN
cana-5407	134	14	process	process	NOUN
cana-5407	134	15	of	of	ADP
cana-5407	134	16	a	a	DET
cana-5407	134	17	machine	machine	NOUN
cana-5407	134	18	learning	learn	VERB
cana-5407	134	19	model	model	NOUN
cana-5407	134	20	.	.	PUNCT
cana-5407	135	1	they	they	PRON
cana-5407	135	2	are	be	AUX
cana-5407	135	3	set	set	VERB
cana-5407	135	4	before	before	ADP
cana-5407	135	5	the	the	DET
cana-5407	135	6	training	training	NOUN
cana-5407	135	7	starts	start	NOUN
cana-5407	135	8	and	and	CCONJ
cana-5407	135	9	can	can	AUX
cana-5407	135	10	have	have	VERB
cana-5407	135	11	a	a	DET
cana-5407	135	12	significant	significant	ADJ
cana-5407	135	13	impact	impact	NOUN
cana-5407	135	14	on	on	ADP
cana-5407	135	15	the	the	DET
cana-5407	135	16	performance	performance	NOUN
cana-5407	135	17	of	of	ADP
cana-5407	135	18	the	the	DET
cana-5407	135	19	model	model	NOUN
cana-5407	135	20	.	.	PUNCT
cana-5407	136	1	the	the	DET
cana-5407	136	2	adjustment	adjustment	NOUN
cana-5407	136	3	of	of	ADP
cana-5407	136	4	hyperparameters	hyperparameter	NOUN
cana-5407	136	5	represents	represent	VERB
cana-5407	136	6	another	another	DET
cana-5407	136	7	level	level	NOUN
cana-5407	136	8	of	of	ADP
cana-5407	136	9	customization	customization	NOUN
cana-5407	136	10	.	.	PUNCT
cana-5407	137	1	hyperparameters	hyperparameter	NOUN
cana-5407	137	2	are	be	AUX
cana-5407	137	3	the	the	DET
cana-5407	137	4	knobs	knob	NOUN
cana-5407	137	5	and	and	CCONJ
cana-5407	137	6	dials	dial	NOUN
cana-5407	137	7	of	of	ADP
cana-5407	137	8	machine	machine	NOUN
cana-5407	137	9	learning	learning	NOUN
cana-5407	137	10	models	model	NOUN
cana-5407	137	11	,	,	PUNCT
cana-5407	137	12	dictating	dictate	VERB
cana-5407	137	13	the	the	DET
cana-5407	137	14	learning	learning	NOUN
cana-5407	137	15	process	process	NOUN
cana-5407	137	16	's	's	PART
cana-5407	137	17	behavior	behavior	NOUN
cana-5407	137	18	.	.	PUNCT
cana-5407	138	1	these	these	DET
cana-5407	138	2	settings	setting	NOUN
cana-5407	138	3	are	be	AUX
cana-5407	138	4	carefully	carefully	ADV
cana-5407	138	5	selected	select	VERB
cana-5407	138	6	to	to	PART
cana-5407	138	7	guide	guide	VERB
cana-5407	138	8	the	the	DET
cana-5407	138	9	learning	learning	NOUN
cana-5407	138	10	process	process	NOUN
cana-5407	138	11	,	,	PUNCT
cana-5407	138	12	potentially	potentially	ADV
cana-5407	138	13	improving	improve	VERB
cana-5407	138	14	the	the	DET
cana-5407	138	15	model	model	NOUN
cana-5407	138	16	's	's	PART
cana-5407	138	17	ability	ability	NOUN
cana-5407	138	18	to	to	PART
cana-5407	138	19	generalize	generalize	VERB
cana-5407	138	20	from	from	ADP
cana-5407	138	21	the	the	DET
cana-5407	138	22	training	training	NOUN
cana-5407	138	23	data	datum	NOUN
cana-5407	138	24	and	and	CCONJ
cana-5407	138	25	ultimately	ultimately	ADV
cana-5407	138	26	affecting	affect	VERB
cana-5407	138	27	the	the	DET
cana-5407	138	28	quality	quality	NOUN
cana-5407	138	29	of	of	ADP
cana-5407	138	30	the	the	DET
cana-5407	138	31	synthesized	synthesize	VERB
cana-5407	138	32	speech	speech	NOUN
cana-5407	138	33	.	.	PUNCT
cana-5407	139	1	several	several	ADJ
cana-5407	139	2	hyperparameters	hyperparameter	NOUN
cana-5407	139	3	were	be	AUX
cana-5407	139	4	adjusted	adjust	VERB
cana-5407	139	5	to	to	PART
cana-5407	139	6	optimize	optimize	VERB
cana-5407	139	7	the	the	DET
cana-5407	139	8	model	model	NOUN
cana-5407	139	9	's	's	PART
cana-5407	139	10	learning	learning	NOUN
cana-5407	139	11	and	and	CCONJ
cana-5407	139	12	generation	generation	NOUN
cana-5407	139	13	capabilities	capability	NOUN
cana-5407	139	14	.	.	PUNCT
cana-5407	140	1	•	•	NUM
cana-5407	140	2	attention	attention	NOUN
cana-5407	140	3	dropout	dropout	NOUN
cana-5407	140	4	rate	rate	NOUN
cana-5407	140	5	:	:	PUNCT
cana-5407	140	6	the	the	DET
cana-5407	140	7	attention	attention	NOUN
cana-5407	140	8	dropout	dropout	NOUN
cana-5407	140	9	rate	rate	NOUN
cana-5407	140	10	is	be	AUX
cana-5407	140	11	a	a	DET
cana-5407	140	12	hyperparameter	hyperparameter	NOUN
cana-5407	140	13	that	that	PRON
cana-5407	140	14	helps	help	VERB
cana-5407	140	15	in	in	ADP
cana-5407	140	16	regularizing	regularize	VERB
cana-5407	140	17	the	the	DET
cana-5407	140	18	model	model	NOUN
cana-5407	140	19	's	's	PART
cana-5407	140	20	training	training	NOUN
cana-5407	140	21	by	by	ADP
cana-5407	140	22	randomly	randomly	ADV
cana-5407	140	23	dropping	drop	VERB
cana-5407	140	24	parts	part	NOUN
cana-5407	140	25	of	of	ADP
cana-5407	140	26	the	the	DET
cana-5407	140	27	data	datum	NOUN
cana-5407	140	28	that	that	PRON
cana-5407	140	29	the	the	DET
cana-5407	140	30	attention	attention	NOUN
cana-5407	140	31	mechanism	mechanism	NOUN
cana-5407	140	32	uses	use	VERB
cana-5407	140	33	to	to	PART
cana-5407	140	34	generate	generate	VERB
cana-5407	140	35	the	the	DET
cana-5407	140	36	context	context	NOUN
cana-5407	140	37	vector	vector	NOUN
cana-5407	140	38	.	.	PUNCT
cana-5407	141	1	this	this	PRON
cana-5407	141	2	is	be	AUX
cana-5407	141	3	a	a	DET
cana-5407	141	4	form	form	NOUN
cana-5407	141	5	of	of	ADP
cana-5407	141	6	dropout	dropout	NOUN
cana-5407	141	7	specifically	specifically	ADV
cana-5407	141	8	applied	apply	VERB
cana-5407	141	9	to	to	ADP
cana-5407	141	10	the	the	DET
cana-5407	141	11	attention	attention	NOUN
cana-5407	141	12	mechanism	mechanism	NOUN
cana-5407	141	13	within	within	ADP
cana-5407	141	14	the	the	DET
cana-5407	141	15	model	model	NOUN
cana-5407	141	16	.	.	PUNCT
cana-5407	142	1	by	by	ADP
cana-5407	142	2	doing	do	VERB
cana-5407	142	3	so	so	ADV
cana-5407	142	4	,	,	PUNCT
cana-5407	142	5	it	it	PRON
cana-5407	142	6	encourages	encourage	VERB
cana-5407	142	7	the	the	DET
cana-5407	142	8	model	model	NOUN
cana-5407	142	9	to	to	PART
cana-5407	142	10	learn	learn	VERB
cana-5407	142	11	robust	robust	ADJ
cana-5407	142	12	representations	representation	NOUN
cana-5407	142	13	without	without	ADP
cana-5407	142	14	overrelying	overrelye	VERB
cana-5407	142	15	on	on	ADP
cana-5407	142	16	specific	specific	ADJ
cana-5407	142	17	parts	part	NOUN
cana-5407	142	18	of	of	ADP
cana-5407	142	19	the	the	DET
cana-5407	142	20	input	input	NOUN
cana-5407	142	21	sequence	sequence	NOUN
cana-5407	142	22	,	,	PUNCT
cana-5407	142	23	potentially	potentially	ADV
cana-5407	142	24	improving	improve	VERB
cana-5407	142	25	the	the	DET
cana-5407	142	26	model	model	NOUN
cana-5407	142	27	's	's	PART
cana-5407	142	28	ability	ability	NOUN
cana-5407	142	29	to	to	PART
cana-5407	142	30	generalize	generalize	VERB
cana-5407	142	31	communications	communication	NOUN
cana-5407	142	32	on	on	ADP
cana-5407	142	33	applied	apply	VERB
cana-5407	142	34	nonlinear	nonlinear	ADJ
cana-5407	142	35	analysis	analysis	NOUN
cana-5407	142	36	issn	issn	NOUN
cana-5407	142	37	:	:	PUNCT
cana-5407	142	38	1074	1074	NUM
cana-5407	142	39	-	-	PUNCT
cana-5407	142	40	133x	133x	NUM
cana-5407	142	41	vol	vol	VERB
cana-5407	142	42	32	32	NUM
cana-5407	142	43	no	no	NOUN
cana-5407	142	44	.	.	PUNCT
cana-5407	143	1	10s	10	NOUN
cana-5407	143	2	(	(	PUNCT
cana-5407	143	3	2025	2025	NUM
cana-5407	143	4	)	)	PUNCT
cana-5407	143	5	2178	2178	NUM
cana-5407	143	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-5407	143	7	and	and	CCONJ
cana-5407	143	8	thus	thus	ADV
cana-5407	143	9	reducing	reduce	VERB
cana-5407	143	10	the	the	DET
cana-5407	143	11	risk	risk	NOUN
cana-5407	143	12	of	of	ADP
cana-5407	143	13	overfitting	overfitte	VERB
cana-5407	143	14	.	.	PUNCT
cana-5407	144	1	it	it	PRON
cana-5407	144	2	's	be	AUX
cana-5407	144	3	a	a	DET
cana-5407	144	4	balancing	balancing	NOUN
cana-5407	144	5	act	act	NOUN
cana-5407	144	6	;	;	PUNCT
cana-5407	144	7	too	too	ADV
cana-5407	144	8	much	much	ADJ
cana-5407	144	9	dropout	dropout	NOUN
cana-5407	144	10	can	can	AUX
cana-5407	144	11	hinder	hinder	VERB
cana-5407	144	12	learning	learn	VERB
cana-5407	144	13	,	,	PUNCT
cana-5407	144	14	while	while	SCONJ
cana-5407	144	15	too	too	ADV
cana-5407	144	16	little	little	ADJ
cana-5407	144	17	can	can	AUX
cana-5407	144	18	lead	lead	VERB
cana-5407	144	19	to	to	ADP
cana-5407	144	20	overfitting	overfitte	VERB
cana-5407	144	21	.	.	PUNCT
cana-5407	145	1	•	•	NUM
cana-5407	145	2	decoder	decoder	NOUN
cana-5407	145	3	dropout	dropout	NOUN
cana-5407	145	4	rate	rate	NOUN
cana-5407	145	5	:	:	PUNCT
cana-5407	145	6	similarly	similarly	ADV
cana-5407	145	7	,	,	PUNCT
cana-5407	145	8	the	the	DET
cana-5407	145	9	decoder	decoder	NOUN
cana-5407	145	10	dropout	dropout	NOUN
cana-5407	145	11	rate	rate	NOUN
cana-5407	145	12	applies	apply	VERB
cana-5407	145	13	the	the	DET
cana-5407	145	14	dropout	dropout	NOUN
cana-5407	145	15	regularization	regularization	NOUN
cana-5407	145	16	technique	technique	NOUN
cana-5407	145	17	to	to	ADP
cana-5407	145	18	the	the	DET
cana-5407	145	19	decoder	decoder	NOUN
cana-5407	145	20	part	part	NOUN
cana-5407	145	21	of	of	ADP
cana-5407	145	22	the	the	DET
cana-5407	145	23	network	network	NOUN
cana-5407	145	24	.	.	PUNCT
cana-5407	146	1	the	the	DET
cana-5407	146	2	decoder	decoder	NOUN
cana-5407	146	3	is	be	AUX
cana-5407	146	4	responsible	responsible	ADJ
cana-5407	146	5	for	for	ADP
cana-5407	146	6	generating	generate	VERB
cana-5407	146	7	the	the	DET
cana-5407	146	8	mel	mel	PROPN
cana-5407	146	9	spectrogram	spectrogram	PROPN
cana-5407	146	10	frame	frame	NOUN
cana-5407	146	11	by	by	ADP
cana-5407	146	12	frame	frame	NOUN
cana-5407	146	13	from	from	ADP
cana-5407	146	14	the	the	DET
cana-5407	146	15	encoded	encode	VERB
cana-5407	146	16	input	input	NOUN
cana-5407	146	17	sequence	sequence	NOUN
cana-5407	146	18	.	.	PUNCT
cana-5407	147	1	applying	apply	VERB
cana-5407	147	2	dropout	dropout	NOUN
cana-5407	147	3	to	to	ADP
cana-5407	147	4	the	the	DET
cana-5407	147	5	decoder	decoder	NOUN
cana-5407	147	6	helps	help	VERB
cana-5407	147	7	prevent	prevent	VERB
cana-5407	147	8	overfitting	overfitte	VERB
cana-5407	147	9	to	to	ADP
cana-5407	147	10	the	the	DET
cana-5407	147	11	training	training	NOUN
cana-5407	147	12	data	datum	NOUN
cana-5407	147	13	,	,	PUNCT
cana-5407	147	14	encouraging	encourage	VERB
cana-5407	147	15	the	the	DET
cana-5407	147	16	model	model	NOUN
cana-5407	147	17	to	to	PART
cana-5407	147	18	find	find	VERB
cana-5407	147	19	more	more	ADJ
cana-5407	147	20	generalized	generalized	ADJ
cana-5407	147	21	solutions	solution	NOUN
cana-5407	147	22	that	that	PRON
cana-5407	147	23	perform	perform	VERB
cana-5407	147	24	better	well	ADV
cana-5407	147	25	on	on	ADP
cana-5407	147	26	unseen	unseen	ADJ
cana-5407	147	27	data	datum	NOUN
cana-5407	147	28	.	.	PUNCT
cana-5407	148	1	just	just	ADV
cana-5407	148	2	as	as	ADP
cana-5407	148	3	with	with	ADP
cana-5407	148	4	attention	attention	NOUN
cana-5407	148	5	dropout	dropout	NOUN
cana-5407	148	6	,	,	PUNCT
cana-5407	148	7	the	the	DET
cana-5407	148	8	decoder	decoder	NOUN
cana-5407	148	9	dropout	dropout	NOUN
cana-5407	148	10	rate	rate	NOUN
cana-5407	148	11	must	must	AUX
cana-5407	148	12	be	be	AUX
cana-5407	148	13	carefully	carefully	ADV
cana-5407	148	14	tuned	tune	VERB
cana-5407	148	15	to	to	PART
cana-5407	148	16	find	find	VERB
cana-5407	148	17	the	the	DET
cana-5407	148	18	right	right	ADJ
cana-5407	148	19	amount	amount	NOUN
cana-5407	148	20	of	of	ADP
cana-5407	148	21	regularization	regularization	NOUN
cana-5407	148	22	.	.	PUNCT
cana-5407	149	1	•	•	NUM
cana-5407	149	2	learning	learn	VERB
cana-5407	149	3	rate	rate	NOUN
cana-5407	149	4	:	:	PUNCT
cana-5407	149	5	the	the	DET
cana-5407	149	6	learning	learning	NOUN
cana-5407	149	7	rate	rate	NOUN
cana-5407	149	8	controls	control	VERB
cana-5407	149	9	how	how	SCONJ
cana-5407	149	10	much	much	ADJ
cana-5407	149	11	the	the	DET
cana-5407	149	12	model	model	NOUN
cana-5407	149	13	's	's	PART
cana-5407	149	14	weights	weight	NOUN
cana-5407	149	15	are	be	AUX
cana-5407	149	16	updated	update	VERB
cana-5407	149	17	during	during	ADP
cana-5407	149	18	training	training	NOUN
cana-5407	149	19	in	in	ADP
cana-5407	149	20	response	response	NOUN
cana-5407	149	21	to	to	ADP
cana-5407	149	22	the	the	DET
cana-5407	149	23	estimated	estimate	VERB
cana-5407	149	24	error	error	NOUN
cana-5407	149	25	each	each	DET
cana-5407	149	26	time	time	NOUN
cana-5407	149	27	the	the	DET
cana-5407	149	28	model	model	NOUN
cana-5407	149	29	weights	weight	NOUN
cana-5407	149	30	are	be	AUX
cana-5407	149	31	updated	update	VERB
cana-5407	149	32	.	.	PUNCT
cana-5407	150	1	it	it	PRON
cana-5407	150	2	is	be	AUX
cana-5407	150	3	one	one	NUM
cana-5407	150	4	of	of	ADP
cana-5407	150	5	the	the	DET
cana-5407	150	6	most	most	ADV
cana-5407	150	7	important	important	ADJ
cana-5407	150	8	hyperparameters	hyperparameter	NOUN
cana-5407	150	9	and	and	CCONJ
cana-5407	150	10	can	can	AUX
cana-5407	150	11	greatly	greatly	ADV
cana-5407	150	12	affect	affect	VERB
cana-5407	150	13	the	the	DET
cana-5407	150	14	model	model	NOUN
cana-5407	150	15	's	's	PART
cana-5407	150	16	performance	performance	NOUN
cana-5407	150	17	and	and	CCONJ
cana-5407	150	18	convergence	convergence	NOUN
cana-5407	150	19	speed	speed	NOUN
cana-5407	150	20	.	.	PUNCT
cana-5407	151	1	a	a	DET
cana-5407	151	2	learning	learning	NOUN
cana-5407	151	3	rate	rate	NOUN
cana-5407	151	4	that	that	PRON
cana-5407	151	5	is	be	AUX
cana-5407	151	6	too	too	ADV
cana-5407	151	7	high	high	ADJ
cana-5407	151	8	may	may	AUX
cana-5407	151	9	cause	cause	VERB
cana-5407	151	10	the	the	DET
cana-5407	151	11	model	model	NOUN
cana-5407	151	12	to	to	PART
cana-5407	151	13	converge	converge	VERB
cana-5407	151	14	too	too	ADV
cana-5407	151	15	quickly	quickly	ADV
cana-5407	151	16	to	to	ADP
cana-5407	151	17	a	a	DET
cana-5407	151	18	suboptimal	suboptimal	ADJ
cana-5407	151	19	solution	solution	NOUN
cana-5407	151	20	,	,	PUNCT
cana-5407	151	21	while	while	SCONJ
cana-5407	151	22	a	a	DET
cana-5407	151	23	learning	learning	NOUN
cana-5407	151	24	rate	rate	NOUN
cana-5407	151	25	that	that	PRON
cana-5407	151	26	is	be	AUX
cana-5407	151	27	too	too	ADV
cana-5407	151	28	low	low	ADJ
cana-5407	151	29	can	can	AUX
cana-5407	151	30	result	result	VERB
cana-5407	151	31	in	in	ADP
cana-5407	151	32	a	a	DET
cana-5407	151	33	long	long	ADJ
cana-5407	151	34	training	training	NOUN
cana-5407	151	35	process	process	NOUN
cana-5407	151	36	that	that	PRON
cana-5407	151	37	could	could	AUX
cana-5407	151	38	stall	stall	VERB
cana-5407	151	39	.	.	PUNCT
cana-5407	152	1	advanced	advanced	ADJ
cana-5407	152	2	techniques	technique	NOUN
cana-5407	152	3	such	such	ADJ
cana-5407	152	4	as	as	ADP
cana-5407	152	5	learning	learn	VERB
cana-5407	152	6	rate	rate	NOUN
cana-5407	152	7	schedules	schedule	NOUN
cana-5407	152	8	or	or	CCONJ
cana-5407	152	9	adaptive	adaptive	ADJ
cana-5407	152	10	learning	learning	NOUN
cana-5407	152	11	rates	rate	NOUN
cana-5407	152	12	can	can	AUX
cana-5407	152	13	be	be	AUX
cana-5407	152	14	used	use	VERB
cana-5407	152	15	to	to	PART
cana-5407	152	16	adjust	adjust	VERB
cana-5407	152	17	the	the	DET
cana-5407	152	18	learning	learning	NOUN
cana-5407	152	19	rate	rate	NOUN
cana-5407	152	20	during	during	ADP
cana-5407	152	21	training	train	VERB
cana-5407	152	22	dynamically	dynamically	ADV
cana-5407	152	23	.	.	PUNCT
cana-5407	153	1	•	•	NOUN
cana-5407	153	2	batch	batch	NOUN
cana-5407	153	3	size	size	NOUN
cana-5407	153	4	:	:	PUNCT
cana-5407	153	5	the	the	DET
cana-5407	153	6	batch	batch	NOUN
cana-5407	153	7	size	size	NOUN
cana-5407	153	8	specifies	specify	VERB
cana-5407	153	9	the	the	DET
cana-5407	153	10	number	number	NOUN
cana-5407	153	11	of	of	ADP
cana-5407	153	12	training	training	NOUN
cana-5407	153	13	samples	sample	NOUN
cana-5407	153	14	to	to	PART
cana-5407	153	15	be	be	AUX
cana-5407	153	16	fed	feed	VERB
cana-5407	153	17	through	through	ADP
cana-5407	153	18	the	the	DET
cana-5407	153	19	network	network	NOUN
cana-5407	153	20	before	before	SCONJ
cana-5407	153	21	the	the	DET
cana-5407	153	22	model	model	NOUN
cana-5407	153	23	's	's	PART
cana-5407	153	24	internal	internal	ADJ
cana-5407	153	25	parameters	parameter	NOUN
cana-5407	153	26	are	be	AUX
cana-5407	153	27	updated	update	VERB
cana-5407	153	28	.	.	PUNCT
cana-5407	154	1	it	it	PRON
cana-5407	154	2	is	be	AUX
cana-5407	154	3	a	a	DET
cana-5407	154	4	compromise	compromise	NOUN
cana-5407	154	5	between	between	ADP
cana-5407	154	6	the	the	DET
cana-5407	154	7	idealized	idealized	ADJ
cana-5407	154	8	stochastic	stochastic	ADJ
cana-5407	154	9	gradient	gradient	ADJ
cana-5407	154	10	descent	descent	NOUN
cana-5407	154	11	(	(	PUNCT
cana-5407	154	12	where	where	SCONJ
cana-5407	154	13	the	the	DET
cana-5407	154	14	batch	batch	NOUN
cana-5407	154	15	size	size	NOUN
cana-5407	154	16	is	be	AUX
cana-5407	154	17	1	1	NUM
cana-5407	154	18	)	)	PUNCT
cana-5407	154	19	and	and	CCONJ
cana-5407	154	20	full	full	ADJ
cana-5407	154	21	batch	batch	NOUN
cana-5407	154	22	training	training	NOUN
cana-5407	154	23	(	(	PUNCT
cana-5407	154	24	where	where	SCONJ
cana-5407	154	25	the	the	DET
cana-5407	154	26	batch	batch	NOUN
cana-5407	154	27	size	size	NOUN
cana-5407	154	28	is	be	AUX
cana-5407	154	29	the	the	DET
cana-5407	154	30	entire	entire	ADJ
cana-5407	154	31	dataset	dataset	NOUN
cana-5407	154	32	)	)	PUNCT
cana-5407	154	33	.	.	PUNCT
cana-5407	155	1	reducing	reduce	VERB
cana-5407	155	2	the	the	DET
cana-5407	155	3	size	size	NOUN
cana-5407	155	4	of	of	ADP
cana-5407	155	5	the	the	DET
cana-5407	155	6	batches	batch	NOUN
cana-5407	155	7	used	use	VERB
cana-5407	155	8	during	during	ADP
cana-5407	155	9	training	training	NOUN
cana-5407	155	10	has	have	VERB
cana-5407	155	11	a	a	DET
cana-5407	155	12	regularization	regularization	NOUN
cana-5407	155	13	impact	impact	NOUN
cana-5407	155	14	and	and	CCONJ
cana-5407	155	15	improves	improve	VERB
cana-5407	155	16	generalization	generalization	NOUN
cana-5407	155	17	.	.	PUNCT
cana-5407	156	1	however	however	ADV
cana-5407	156	2	,	,	PUNCT
cana-5407	156	3	this	this	PRON
cana-5407	156	4	may	may	AUX
cana-5407	156	5	introduce	introduce	VERB
cana-5407	156	6	more	more	ADJ
cana-5407	156	7	noise	noise	NOUN
cana-5407	156	8	and	and	CCONJ
cana-5407	156	9	lengthen	lengthen	VERB
cana-5407	156	10	the	the	DET
cana-5407	156	11	training	training	NOUN
cana-5407	156	12	process	process	NOUN
cana-5407	156	13	.	.	PUNCT
cana-5407	157	1	increasing	increase	VERB
cana-5407	157	2	the	the	DET
cana-5407	157	3	size	size	NOUN
cana-5407	157	4	of	of	ADP
cana-5407	157	5	the	the	DET
cana-5407	157	6	batches	batch	NOUN
cana-5407	157	7	used	use	VERB
cana-5407	157	8	in	in	ADP
cana-5407	157	9	training	training	NOUN
cana-5407	157	10	may	may	AUX
cana-5407	157	11	provide	provide	VERB
cana-5407	157	12	more	more	ADV
cana-5407	157	13	precise	precise	ADJ
cana-5407	157	14	gradient	gradient	NOUN
cana-5407	157	15	estimates	estimate	NOUN
cana-5407	157	16	,	,	PUNCT
cana-5407	157	17	but	but	CCONJ
cana-5407	157	18	it	it	PRON
cana-5407	157	19	may	may	AUX
cana-5407	157	20	also	also	ADV
cana-5407	157	21	need	need	VERB
cana-5407	157	22	a	a	DET
cana-5407	157	23	larger	large	ADJ
cana-5407	157	24	amount	amount	NOUN
cana-5407	157	25	of	of	ADP
cana-5407	157	26	memory	memory	NOUN
cana-5407	157	27	and	and	CCONJ
cana-5407	157	28	perhaps	perhaps	ADV
cana-5407	157	29	result	result	VERB
cana-5407	157	30	in	in	ADP
cana-5407	157	31	faster	fast	ADJ
cana-5407	157	32	convergence	convergence	NOUN
cana-5407	157	33	towards	towards	ADP
cana-5407	157	34	inferior	inferior	ADJ
cana-5407	157	35	minimum	minimum	ADJ
cana-5407	157	36	points	point	NOUN
cana-5407	157	37	.	.	PUNCT
cana-5407	158	1	3.3	3.3	NUM
cana-5407	158	2	pronunciation	pronunciation	NOUN
cana-5407	158	3	and	and	CCONJ
cana-5407	158	4	language	language	NOUN
cana-5407	158	5	handling	handle	VERB
cana-5407	158	6	accurate	accurate	ADJ
cana-5407	158	7	and	and	CCONJ
cana-5407	158	8	natural	natural	ADJ
cana-5407	158	9	-	-	PUNCT
cana-5407	158	10	sounding	sound	VERB
cana-5407	158	11	pronunciation	pronunciation	NOUN
cana-5407	158	12	is	be	AUX
cana-5407	158	13	just	just	ADV
cana-5407	158	14	as	as	ADV
cana-5407	158	15	crucial	crucial	ADJ
cana-5407	158	16	in	in	ADP
cana-5407	158	17	voice	voice	NOUN
cana-5407	158	18	synthesis	synthesis	NOUN
cana-5407	158	19	systems	system	NOUN
cana-5407	158	20	to	to	PART
cana-5407	158	21	provide	provide	VERB
cana-5407	158	22	a	a	DET
cana-5407	158	23	lifelike	lifelike	ADJ
cana-5407	158	24	and	and	CCONJ
cana-5407	158	25	enticing	enticing	ADJ
cana-5407	158	26	listening	listening	NOUN
cana-5407	158	27	experience	experience	NOUN
cana-5407	158	28	.	.	PUNCT
cana-5407	159	1	an	an	DET
cana-5407	159	2	especially	especially	ADV
cana-5407	159	3	new	new	ADJ
cana-5407	159	4	technique	technique	NOUN
cana-5407	159	5	for	for	ADP
cana-5407	159	6	influencing	influence	VERB
cana-5407	159	7	pronunciation	pronunciation	NOUN
cana-5407	159	8	in	in	ADP
cana-5407	159	9	text	text	NOUN
cana-5407	159	10	-	-	PUNCT
cana-5407	159	11	to	to	ADP
cana-5407	159	12	-	-	PUNCT
cana-5407	159	13	speech	speech	NOUN
cana-5407	159	14	systems	system	NOUN
cana-5407	159	15	is	be	AUX
cana-5407	159	16	to	to	PART
cana-5407	159	17	utilize	utilize	VERB
cana-5407	159	18	the	the	DET
cana-5407	159	19	arpabet	arpabet	ADJ
cana-5407	159	20	phoneme	phoneme	NOUN
cana-5407	159	21	set	set	NOUN
cana-5407	159	22	to	to	PART
cana-5407	159	23	transform	transform	VERB
cana-5407	159	24	typical	typical	ADJ
cana-5407	159	25	written	write	VERB
cana-5407	159	26	text	text	NOUN
cana-5407	159	27	into	into	ADP
cana-5407	159	28	intellectual	intellectual	ADJ
cana-5407	159	29	transcript	transcript	NOUN
cana-5407	159	30	phonetics	phonetic	NOUN
cana-5407	159	31	based	base	VERB
cana-5407	159	32	on	on	ADP
cana-5407	159	33	a	a	DET
cana-5407	159	34	dictionary	dictionary	NOUN
cana-5407	159	35	.	.	PUNCT
cana-5407	160	1	the	the	DET
cana-5407	160	2	primary	primary	ADJ
cana-5407	160	3	purpose	purpose	NOUN
cana-5407	160	4	of	of	ADP
cana-5407	160	5	this	this	DET
cana-5407	160	6	approach	approach	NOUN
cana-5407	160	7	is	be	AUX
cana-5407	160	8	to	to	PART
cana-5407	160	9	provide	provide	VERB
cana-5407	160	10	a	a	DET
cana-5407	160	11	unique	unique	ADJ
cana-5407	160	12	and	and	CCONJ
cana-5407	160	13	precise	precise	ADJ
cana-5407	160	14	set	set	NOUN
cana-5407	160	15	of	of	ADP
cana-5407	160	16	pronunciation	pronunciation	NOUN
cana-5407	160	17	instructions	instruction	NOUN
cana-5407	160	18	for	for	ADP
cana-5407	160	19	the	the	DET
cana-5407	160	20	voice	voice	NOUN
cana-5407	160	21	synthesis	synthesis	NOUN
cana-5407	160	22	engine	engine	NOUN
cana-5407	160	23	.	.	PUNCT
cana-5407	161	1	arpabet	arpabet	PROPN
cana-5407	161	2	is	be	AUX
cana-5407	161	3	a	a	DET
cana-5407	161	4	set	set	NOUN
cana-5407	161	5	of	of	ADP
cana-5407	161	6	pronunciation	pronunciation	NOUN
cana-5407	161	7	transcription	transcription	NOUN
cana-5407	161	8	codes	code	NOUN
cana-5407	161	9	that	that	PRON
cana-5407	161	10	are	be	AUX
cana-5407	161	11	commonly	commonly	ADV
cana-5407	161	12	utilized	utilize	VERB
cana-5407	161	13	for	for	ADP
cana-5407	161	14	voice	voice	NOUN
cana-5407	161	15	and	and	CCONJ
cana-5407	161	16	speech	speech	NOUN
cana-5407	161	17	processing	processing	NOUN
cana-5407	161	18	purposes	purpose	NOUN
cana-5407	161	19	.	.	PUNCT
cana-5407	162	1	they	they	PRON
cana-5407	162	2	are	be	AUX
cana-5407	162	3	symbols	symbol	NOUN
cana-5407	162	4	or	or	CCONJ
cana-5407	162	5	a	a	DET
cana-5407	162	6	combination	combination	NOUN
cana-5407	162	7	of	of	ADP
cana-5407	162	8	symbols	symbol	NOUN
cana-5407	162	9	that	that	PRON
cana-5407	162	10	stand	stand	VERB
cana-5407	162	11	for	for	ADP
cana-5407	162	12	each	each	DET
cana-5407	162	13	phoneme	phoneme	NOUN
cana-5407	162	14	in	in	ADP
cana-5407	162	15	the	the	DET
cana-5407	162	16	english	english	ADJ
cana-5407	162	17	language	language	NOUN
cana-5407	162	18	.	.	PUNCT
cana-5407	163	1	phonetic	phonetic	ADJ
cana-5407	163	2	rendering	rendering	NOUN
cana-5407	163	3	is	be	AUX
cana-5407	163	4	critical	critical	ADJ
cana-5407	163	5	since	since	SCONJ
cana-5407	163	6	it	it	PRON
cana-5407	163	7	enables	enable	VERB
cana-5407	163	8	the	the	DET
cana-5407	163	9	exact	exact	ADJ
cana-5407	163	10	meeting	meeting	NOUN
cana-5407	163	11	of	of	ADP
cana-5407	163	12	the	the	DET
cana-5407	163	13	sounds	sound	NOUN
cana-5407	163	14	that	that	PRON
cana-5407	163	15	constitute	constitute	VERB
cana-5407	163	16	spoken	speak	VERB
cana-5407	163	17	words	word	NOUN
cana-5407	163	18	to	to	PART
cana-5407	163	19	be	be	AUX
cana-5407	163	20	appropriately	appropriately	ADV
cana-5407	163	21	determined	determine	VERB
cana-5407	163	22	rather	rather	ADV
cana-5407	163	23	than	than	ADP
cana-5407	163	24	relying	rely	VERB
cana-5407	163	25	on	on	ADP
cana-5407	163	26	the	the	DET
cana-5407	163	27	infrequent	infrequent	ADJ
cana-5407	163	28	and	and	CCONJ
cana-5407	163	29	inconsistent	inconsistent	ADJ
cana-5407	163	30	pronunciation	pronunciation	NOUN
cana-5407	163	31	based	base	VERB
cana-5407	163	32	on	on	ADP
cana-5407	163	33	the	the	DET
cana-5407	163	34	standard	standard	ADJ
cana-5407	163	35	manuscript	manuscript	NOUN
cana-5407	163	36	.	.	PUNCT
cana-5407	164	1	for	for	ADP
cana-5407	164	2	instance	instance	NOUN
cana-5407	164	3	,	,	PUNCT
cana-5407	164	4	a	a	DET
cana-5407	164	5	word	word	NOUN
cana-5407	164	6	such	such	ADJ
cana-5407	164	7	as	as	ADP
cana-5407	164	8	“	"	PUNCT
cana-5407	164	9	enough	enough	ADJ
cana-5407	164	10	”	"	PUNCT
cana-5407	164	11	may	may	AUX
cana-5407	164	12	be	be	AUX
cana-5407	164	13	pronounced	pronounce	VERB
cana-5407	164	14	in	in	ADP
cana-5407	164	15	various	various	ADJ
cana-5407	164	16	ways	way	NOUN
cana-5407	164	17	depending	depend	VERB
cana-5407	164	18	on	on	ADP
cana-5407	164	19	how	how	SCONJ
cana-5407	164	20	it	it	PRON
cana-5407	164	21	is	be	AUX
cana-5407	164	22	spelt	spelt	NOUN
cana-5407	164	23	while	while	SCONJ
cana-5407	164	24	offering	offer	VERB
cana-5407	164	25	the	the	DET
cana-5407	164	26	same	same	ADJ
cana-5407	164	27	transcription	transcription	NOUN
cana-5407	164	28	an	an	DET
cana-5407	164	29	arpabet	arpabet	ADJ
cana-5407	164	30	symbol	symbol	NOUN
cana-5407	164	31	generates	generate	VERB
cana-5407	164	32	an	an	DET
cana-5407	164	33	exact	exact	ADJ
cana-5407	164	34	reference	reference	NOUN
cana-5407	164	35	to	to	ADP
cana-5407	164	36	the	the	DET
cana-5407	164	37	pronunciation	pronunciation	NOUN
cana-5407	164	38	.	.	PUNCT
cana-5407	165	1	a	a	DET
cana-5407	165	2	dictionary	dictionary	ADJ
cana-5407	165	3	-	-	PUNCT
cana-5407	165	4	based	base	VERB
cana-5407	165	5	pronunciation	pronunciation	NOUN
cana-5407	165	6	system	system	NOUN
cana-5407	165	7	uses	use	VERB
cana-5407	165	8	a	a	DET
cana-5407	165	9	massive	massive	ADJ
cana-5407	165	10	database	database	NOUN
cana-5407	165	11	.	.	PUNCT
cana-5407	166	1	in	in	ADP
cana-5407	166	2	this	this	DET
cana-5407	166	3	database	database	NOUN
cana-5407	166	4	,	,	PUNCT
cana-5407	166	5	words	word	NOUN
cana-5407	166	6	are	be	AUX
cana-5407	166	7	associated	associate	VERB
cana-5407	166	8	with	with	ADP
cana-5407	166	9	short	short	ADJ
cana-5407	166	10	arpabet	arpabet	ADJ
cana-5407	166	11	transcriptions	transcription	NOUN
cana-5407	166	12	that	that	PRON
cana-5407	166	13	portray	portray	VERB
cana-5407	166	14	their	their	PRON
cana-5407	166	15	pronunciation	pronunciation	NOUN
cana-5407	166	16	.	.	PUNCT
cana-5407	167	1	while	while	SCONJ
cana-5407	167	2	text	text	NOUN
cana-5407	167	3	is	be	AUX
cana-5407	167	4	being	be	AUX
cana-5407	167	5	spoken	speak	VERB
cana-5407	167	6	in	in	ADP
cana-5407	167	7	the	the	DET
cana-5407	167	8	given	give	VERB
cana-5407	167	9	tts	tts	PROPN
cana-5407	167	10	system	system	NOUN
cana-5407	167	11	,	,	PUNCT
cana-5407	167	12	the	the	DET
cana-5407	167	13	dictionary	dictionary	NOUN
cana-5407	167	14	is	be	AUX
cana-5407	167	15	consulted	consult	VERB
cana-5407	167	16	in	in	ADP
cana-5407	167	17	reverse	reverse	NOUN
cana-5407	167	18	to	to	PART
cana-5407	167	19	produce	produce	VERB
cana-5407	167	20	the	the	DET
cana-5407	167	21	transcription	transcription	NOUN
cana-5407	167	22	communications	communication	NOUN
cana-5407	167	23	on	on	ADP
cana-5407	167	24	applied	apply	VERB
cana-5407	167	25	nonlinear	nonlinear	ADJ
cana-5407	167	26	analysis	analysis	NOUN
cana-5407	167	27	issn	issn	NOUN
cana-5407	167	28	:	:	PUNCT
cana-5407	167	29	1074	1074	NUM
cana-5407	167	30	-	-	PUNCT
cana-5407	167	31	133x	133x	NUM
cana-5407	167	32	vol	vol	VERB
cana-5407	167	33	32	32	NUM
cana-5407	167	34	no	no	NOUN
cana-5407	167	35	.	.	PUNCT
cana-5407	168	1	10s	10	NOUN
cana-5407	168	2	(	(	PUNCT
cana-5407	168	3	2025	2025	NUM
cana-5407	168	4	)	)	PUNCT
cana-5407	168	5	2179	2179	NUM
cana-5407	168	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-5407	168	7	for	for	ADP
cana-5407	168	8	each	each	DET
cana-5407	168	9	word	word	NOUN
cana-5407	168	10	.	.	PUNCT
cana-5407	169	1	this	this	PRON
cana-5407	169	2	ensures	ensure	VERB
cana-5407	169	3	that	that	SCONJ
cana-5407	169	4	the	the	DET
cana-5407	169	5	synthesis	synthesis	NOUN
cana-5407	169	6	engine	engine	NOUN
cana-5407	169	7	definitively	definitively	ADV
cana-5407	169	8	knows	know	VERB
cana-5407	169	9	how	how	SCONJ
cana-5407	169	10	to	to	PART
cana-5407	169	11	properly	properly	ADV
cana-5407	169	12	express	express	VERB
cana-5407	169	13	each	each	DET
cana-5407	169	14	word	word	NOUN
cana-5407	169	15	,	,	PUNCT
cana-5407	169	16	regardless	regardless	ADV
cana-5407	169	17	of	of	ADP
cana-5407	169	18	context	context	NOUN
cana-5407	169	19	or	or	CCONJ
cana-5407	169	20	the	the	DET
cana-5407	169	21	spellings	spelling	NOUN
cana-5407	169	22	of	of	ADP
cana-5407	169	23	the	the	DET
cana-5407	169	24	words	word	NOUN
cana-5407	169	25	surrounding	surround	VERB
cana-5407	169	26	it	it	PRON
cana-5407	169	27	.	.	PUNCT
cana-5407	170	1	the	the	DET
cana-5407	170	2	advantages	advantage	NOUN
cana-5407	170	3	of	of	ADP
cana-5407	170	4	using	use	VERB
cana-5407	170	5	a	a	DET
cana-5407	170	6	pronunciation	pronunciation	NOUN
cana-5407	170	7	dictionary	dictionary	NOUN
cana-5407	170	8	in	in	ADP
cana-5407	170	9	speech	speech	NOUN
cana-5407	170	10	synthesis	synthesis	NOUN
cana-5407	170	11	are	be	AUX
cana-5407	170	12	as	as	SCONJ
cana-5407	170	13	follows	follow	VERB
cana-5407	170	14	:	:	PUNCT
cana-5407	170	15	•	•	NUM
cana-5407	170	16	consistency	consistency	NOUN
cana-5407	170	17	:	:	PUNCT
cana-5407	170	18	pronunciation	pronunciation	NOUN
cana-5407	170	19	is	be	AUX
cana-5407	170	20	consistent	consistent	ADJ
cana-5407	170	21	between	between	ADP
cana-5407	170	22	different	different	ADJ
cana-5407	170	23	instances	instance	NOUN
cana-5407	170	24	of	of	ADP
cana-5407	170	25	the	the	DET
cana-5407	170	26	same	same	ADJ
cana-5407	170	27	word	word	NOUN
cana-5407	170	28	,	,	PUNCT
cana-5407	170	29	making	make	VERB
cana-5407	170	30	speech	speech	NOUN
cana-5407	170	31	output	output	NOUN
cana-5407	170	32	more	more	ADJ
cana-5407	170	33	uniform	uniform	ADJ
cana-5407	170	34	and	and	CCONJ
cana-5407	170	35	predictable	predictable	ADJ
cana-5407	170	36	.	.	PUNCT
cana-5407	171	1	•	•	NUM
cana-5407	171	2	accuracy	accuracy	NOUN
cana-5407	171	3	:	:	PUNCT
cana-5407	171	4	words	word	NOUN
cana-5407	171	5	that	that	PRON
cana-5407	171	6	are	be	AUX
cana-5407	171	7	either	either	ADV
cana-5407	171	8	difficult	difficult	ADJ
cana-5407	171	9	to	to	PART
cana-5407	171	10	pronounce	pronounce	VERB
cana-5407	171	11	,	,	PUNCT
cana-5407	171	12	or	or	CCONJ
cana-5407	171	13	out	out	ADV
cana-5407	171	14	-	-	PUNCT
cana-5407	171	15	of	of	ADP
cana-5407	171	16	-	-	PUNCT
cana-5407	171	17	dictionary	dictionary	NOUN
cana-5407	171	18	and	and	CCONJ
cana-5407	171	19	thus	thus	ADV
cana-5407	171	20	easily	easily	ADV
cana-5407	171	21	mispronounced	mispronounce	VERB
cana-5407	171	22	due	due	ADP
cana-5407	171	23	to	to	ADP
cana-5407	171	24	their	their	PRON
cana-5407	171	25	irregular	irregular	ADJ
cana-5407	171	26	spelling	spelling	NOUN
cana-5407	171	27	and	and	CCONJ
cana-5407	171	28	are	be	AUX
cana-5407	171	29	pronounced	pronounce	VERB
cana-5407	171	30	regarding	regard	VERB
cana-5407	171	31	their	their	PRON
cana-5407	171	32	contexts	context	NOUN
cana-5407	171	33	,	,	PUNCT
cana-5407	171	34	are	be	AUX
cana-5407	171	35	pronounced	pronounce	VERB
cana-5407	171	36	correctly	correctly	ADV
cana-5407	171	37	.	.	PUNCT
cana-5407	172	1	this	this	PRON
cana-5407	172	2	can	can	AUX
cana-5407	172	3	lead	lead	VERB
cana-5407	172	4	to	to	ADP
cana-5407	172	5	an	an	DET
cana-5407	172	6	overall	overall	ADJ
cana-5407	172	7	increase	increase	NOUN
cana-5407	172	8	in	in	ADP
cana-5407	172	9	the	the	DET
cana-5407	172	10	naturalness	naturalness	NOUN
cana-5407	172	11	and	and	CCONJ
cana-5407	172	12	intelligibility	intelligibility	NOUN
cana-5407	172	13	of	of	ADP
cana-5407	172	14	a	a	DET
cana-5407	172	15	synthesized	synthesize	VERB
cana-5407	172	16	text	text	NOUN
cana-5407	172	17	.	.	PUNCT
cana-5407	173	1	•	•	NUM
cana-5407	173	2	flexibility	flexibility	NOUN
cana-5407	173	3	:	:	PUNCT
cana-5407	173	4	the	the	DET
cana-5407	173	5	correct	correct	ADJ
cana-5407	173	6	pronunciation	pronunciation	NOUN
cana-5407	173	7	of	of	ADP
cana-5407	173	8	new	new	ADJ
cana-5407	173	9	words	word	NOUN
cana-5407	173	10	or	or	CCONJ
cana-5407	173	11	special	special	ADJ
cana-5407	173	12	knowledge	knowledge	NOUN
cana-5407	173	13	such	such	ADJ
cana-5407	173	14	as	as	ADP
cana-5407	173	15	medical	medical	ADJ
cana-5407	173	16	terms	term	NOUN
cana-5407	173	17	,	,	PUNCT
cana-5407	173	18	or	or	CCONJ
cana-5407	173	19	new	new	ADJ
cana-5407	173	20	names	name	NOUN
cana-5407	173	21	,	,	PUNCT
cana-5407	173	22	may	may	AUX
cana-5407	173	23	be	be	AUX
cana-5407	173	24	added	add	VERB
cana-5407	173	25	to	to	ADP
cana-5407	173	26	the	the	DET
cana-5407	173	27	dictionary	dictionary	PROPN
cana-5407	173	28	.	.	PUNCT
cana-5407	174	1	this	this	DET
cana-5407	174	2	flexibility	flexibility	NOUN
cana-5407	174	3	allows	allow	VERB
cana-5407	174	4	the	the	DET
cana-5407	174	5	system	system	NOUN
cana-5407	174	6	to	to	PART
cana-5407	174	7	adapt	adapt	VERB
cana-5407	174	8	to	to	ADP
cana-5407	174	9	new	new	ADJ
cana-5407	174	10	vocabularies	vocabulary	NOUN
cana-5407	174	11	,	,	PUNCT
cana-5407	174	12	even	even	ADV
cana-5407	174	13	though	though	SCONJ
cana-5407	174	14	such	such	ADJ
cana-5407	174	15	systems	system	NOUN
cana-5407	174	16	do	do	AUX
cana-5407	174	17	not	not	PART
cana-5407	174	18	require	require	VERB
cana-5407	174	19	retraining	retrain	VERB
cana-5407	174	20	.	.	PUNCT
cana-5407	175	1	•	•	NUM
cana-5407	175	2	efficiency	efficiency	NOUN
cana-5407	175	3	:	:	PUNCT
cana-5407	175	4	because	because	SCONJ
cana-5407	175	5	text	text	NOUN
cana-5407	175	6	-	-	PUNCT
cana-5407	175	7	to	to	ADP
cana-5407	175	8	-	-	PUNCT
cana-5407	175	9	phoneme	phoneme	NOUN
cana-5407	175	10	conversion	conversion	NOUN
cana-5407	175	11	occurs	occur	VERB
cana-5407	175	12	before	before	ADP
cana-5407	175	13	synthesis	synthesis	NOUN
cana-5407	175	14	,	,	PUNCT
cana-5407	175	15	the	the	DET
cana-5407	175	16	tts	tts	PROPN
cana-5407	175	17	engine	engine	NOUN
cana-5407	175	18	may	may	AUX
cana-5407	175	19	devote	devote	VERB
cana-5407	175	20	more	more	ADV
cana-5407	175	21	computational	computational	ADJ
cana-5407	175	22	attention	attention	NOUN
cana-5407	175	23	to	to	ADP
cana-5407	175	24	human	human	NOUN
cana-5407	175	25	-	-	PUNCT
cana-5407	175	26	like	like	ADJ
cana-5407	175	27	vocoding	vocoding	NOUN
cana-5407	175	28	.	.	PUNCT
cana-5407	176	1	this	this	PRON
cana-5407	176	2	simplifies	simplify	VERB
cana-5407	176	3	the	the	DET
cana-5407	176	4	task	task	NOUN
cana-5407	176	5	of	of	ADP
cana-5407	176	6	synthesis	synthesis	NOUN
cana-5407	176	7	by	by	ADP
cana-5407	176	8	removing	remove	VERB
cana-5407	176	9	the	the	DET
cana-5407	176	10	need	need	NOUN
cana-5407	176	11	to	to	PART
cana-5407	176	12	ascertain	ascertain	VERB
cana-5407	176	13	proper	proper	ADJ
cana-5407	176	14	pronunciation	pronunciation	NOUN
cana-5407	176	15	from	from	ADP
cana-5407	176	16	context	context	NOUN
cana-5407	176	17	.	.	PUNCT
cana-5407	177	1	3.4	3.4	NUM
cana-5407	177	2	output	output	NOUN
cana-5407	177	3	audio	audio	NOUN
cana-5407	177	4	processing	processing	NOUN
cana-5407	177	5	using	use	VERB
cana-5407	177	6	hifi	hifi	NOUN
cana-5407	177	7	-	-	PUNCT
cana-5407	177	8	gan	gan	NOUN
cana-5407	177	9	in	in	ADP
cana-5407	177	10	the	the	DET
cana-5407	177	11	context	context	NOUN
cana-5407	177	12	of	of	ADP
cana-5407	177	13	text	text	NOUN
cana-5407	177	14	-	-	PUNCT
cana-5407	177	15	to	to	ADP
cana-5407	177	16	-	-	PUNCT
cana-5407	177	17	speech	speech	NOUN
cana-5407	177	18	(	(	PUNCT
cana-5407	177	19	tts	tts	NOUN
cana-5407	177	20	)	)	PUNCT
cana-5407	177	21	synthesis	synthesis	NOUN
cana-5407	177	22	,	,	PUNCT
cana-5407	177	23	advanced	advanced	ADJ
cana-5407	177	24	audio	audio	NOUN
cana-5407	177	25	processing	processing	NOUN
cana-5407	177	26	enables	enable	VERB
cana-5407	177	27	one	one	NUM
cana-5407	177	28	to	to	PART
cana-5407	177	29	convert	convert	VERB
cana-5407	177	30	a	a	DET
cana-5407	177	31	simple	simple	ADJ
cana-5407	177	32	text	text	NOUN
cana-5407	177	33	input	input	NOUN
cana-5407	177	34	into	into	ADP
cana-5407	177	35	a	a	DET
cana-5407	177	36	voice	voice	NOUN
cana-5407	177	37	that	that	PRON
cana-5407	177	38	sounds	sound	VERB
cana-5407	177	39	almost	almost	ADV
cana-5407	177	40	natural	natural	ADJ
cana-5407	177	41	.	.	PUNCT
cana-5407	178	1	moreover	moreover	ADV
cana-5407	178	2	,	,	PUNCT
cana-5407	178	3	depending	depend	VERB
cana-5407	178	4	on	on	ADP
cana-5407	178	5	the	the	DET
cana-5407	178	6	level	level	NOUN
cana-5407	178	7	of	of	ADP
cana-5407	178	8	its	its	PRON
cana-5407	178	9	sophistication	sophistication	NOUN
cana-5407	178	10	,	,	PUNCT
cana-5407	178	11	the	the	DET
cana-5407	178	12	quality	quality	NOUN
cana-5407	178	13	,	,	PUNCT
cana-5407	178	14	degree	degree	NOUN
cana-5407	178	15	of	of	ADP
cana-5407	178	16	clarity	clarity	NOUN
cana-5407	178	17	,	,	PUNCT
cana-5407	178	18	and	and	CCONJ
cana-5407	178	19	the	the	DET
cana-5407	178	20	naturalness	naturalness	NOUN
cana-5407	178	21	of	of	ADP
cana-5407	178	22	such	such	ADJ
cana-5407	178	23	speech	speech	NOUN
cana-5407	178	24	are	be	AUX
cana-5407	178	25	achieved	achieve	VERB
cana-5407	178	26	.	.	PUNCT
cana-5407	179	1	in	in	ADP
cana-5407	179	2	this	this	DET
cana-5407	179	3	case	case	NOUN
cana-5407	179	4	,	,	PUNCT
cana-5407	179	5	we	we	PRON
cana-5407	179	6	obtain	obtain	VERB
cana-5407	179	7	the	the	DET
cana-5407	179	8	neural	neural	ADJ
cana-5407	179	9	network	network	NOUN
cana-5407	179	10	models	model	NOUN
cana-5407	179	11	that	that	PRON
cana-5407	179	12	synthesize	synthesize	VERB
cana-5407	179	13	speech	speech	NOUN
cana-5407	179	14	from	from	ADP
cana-5407	179	15	our	our	PRON
cana-5407	179	16	input	input	NOUN
cana-5407	179	17	audio	audio	NOUN
cana-5407	179	18	.	.	PUNCT
cana-5407	180	1	then	then	ADV
cana-5407	180	2	,	,	PUNCT
cana-5407	180	3	increase	increase	VERB
cana-5407	180	4	the	the	DET
cana-5407	180	5	audio	audio	ADJ
cana-5407	180	6	quality	quality	NOUN
cana-5407	180	7	using	use	VERB
cana-5407	180	8	models	model	NOUN
cana-5407	180	9	such	such	ADJ
cana-5407	180	10	as	as	ADP
cana-5407	180	11	hifi	hifi	NOUN
cana-5407	180	12	-	-	PUNCT
cana-5407	180	13	gan	gan	PROPN
cana-5407	180	14	,	,	PUNCT
cana-5407	180	15	which	which	PRON
cana-5407	180	16	follows	follow	VERB
cana-5407	180	17	with	with	ADP
cana-5407	180	18	a	a	DET
cana-5407	180	19	super	super	ADJ
cana-5407	180	20	resolution	resolution	NOUN
cana-5407	180	21	mechanism	mechanism	NOUN
cana-5407	180	22	to	to	PART
cana-5407	180	23	make	make	VERB
cana-5407	180	24	it	it	PRON
cana-5407	180	25	even	even	ADV
cana-5407	180	26	more	more	ADV
cana-5407	180	27	accurate	accurate	ADJ
cana-5407	180	28	.	.	PUNCT
cana-5407	181	1	it	it	PRON
cana-5407	181	2	achieves	achieve	VERB
cana-5407	181	3	this	this	PRON
cana-5407	181	4	by	by	ADP
cana-5407	181	5	generating	generate	VERB
cana-5407	181	6	a	a	DET
cana-5407	181	7	mel	mel	PROPN
cana-5407	181	8	spectrogram	spectrogram	NOUN
cana-5407	181	9	,	,	PUNCT
cana-5407	181	10	a	a	DET
cana-5407	181	11	representation	representation	NOUN
cana-5407	181	12	of	of	ADP
cana-5407	181	13	the	the	DET
cana-5407	181	14	spectrum	spectrum	NOUN
cana-5407	181	15	of	of	ADP
cana-5407	181	16	frequency	frequency	NOUN
cana-5407	181	17	of	of	ADP
cana-5407	181	18	sound	sound	NOUN
cana-5407	181	19	as	as	SCONJ
cana-5407	181	20	they	they	PRON
cana-5407	181	21	change	change	VERB
cana-5407	181	22	over	over	ADP
cana-5407	181	23	time	time	NOUN
cana-5407	181	24	from	from	ADP
cana-5407	181	25	the	the	DET
cana-5407	181	26	text	text	NOUN
cana-5407	181	27	.	.	PUNCT
cana-5407	182	1	this	this	DET
cana-5407	182	2	stage	stage	NOUN
cana-5407	182	3	is	be	AUX
cana-5407	182	4	critical	critical	ADJ
cana-5407	182	5	since	since	SCONJ
cana-5407	182	6	the	the	DET
cana-5407	182	7	mel	mel	PROPN
cana-5407	182	8	spectrogram	spectrogram	PROPN
cana-5407	182	9	adequately	adequately	ADV
cana-5407	182	10	captures	capture	VERB
cana-5407	182	11	the	the	DET
cana-5407	182	12	dynamics	dynamic	NOUN
cana-5407	182	13	and	and	CCONJ
cana-5407	182	14	intonation	intonation	NOUN
cana-5407	182	15	present	present	ADJ
cana-5407	182	16	in	in	ADP
cana-5407	182	17	the	the	DET
cana-5407	182	18	speech	speech	NOUN
cana-5407	182	19	,	,	PUNCT
cana-5407	182	20	being	be	AUX
cana-5407	182	21	a	a	DET
cana-5407	182	22	link	link	NOUN
cana-5407	182	23	between	between	ADP
cana-5407	182	24	raw	raw	ADJ
cana-5407	182	25	audio	audio	NOUN
cana-5407	182	26	and	and	CCONJ
cana-5407	182	27	text	text	NOUN
cana-5407	182	28	and	and	CCONJ
cana-5407	182	29	retaining	retain	VERB
cana-5407	182	30	the	the	DET
cana-5407	182	31	natural	natural	ADJ
cana-5407	182	32	emotional	emotional	ADJ
cana-5407	182	33	and	and	CCONJ
cana-5407	182	34	tonal	tonal	ADJ
cana-5407	182	35	flow	flow	NOUN
cana-5407	182	36	of	of	ADP
cana-5407	182	37	speech	speech	NOUN
cana-5407	182	38	.	.	PUNCT
cana-5407	183	1	upon	upon	SCONJ
cana-5407	183	2	generation	generation	NOUN
cana-5407	183	3	of	of	ADP
cana-5407	183	4	the	the	DET
cana-5407	183	5	mel	mel	PROPN
cana-5407	183	6	spectrogram	spectrogram	PROPN
cana-5407	183	7	,	,	PUNCT
cana-5407	183	8	the	the	DET
cana-5407	183	9	next	next	ADJ
cana-5407	183	10	step	step	NOUN
cana-5407	183	11	is	be	AUX
cana-5407	183	12	to	to	PART
cana-5407	183	13	convert	convert	VERB
cana-5407	183	14	the	the	DET
cana-5407	183	15	aforementioned	aforementioned	ADJ
cana-5407	183	16	frequencytime	frequencytime	NOUN
cana-5407	183	17	graph	graph	NOUN
cana-5407	183	18	into	into	ADP
cana-5407	183	19	waveforms	waveform	NOUN
cana-5407	183	20	that	that	PRON
cana-5407	183	21	can	can	AUX
cana-5407	183	22	be	be	AUX
cana-5407	183	23	heard	hear	VERB
cana-5407	183	24	.	.	PUNCT
cana-5407	184	1	this	this	PRON
cana-5407	184	2	is	be	AUX
cana-5407	184	3	the	the	DET
cana-5407	184	4	role	role	NOUN
cana-5407	184	5	that	that	PRON
cana-5407	184	6	hifi	hifi	PROPN
cana-5407	184	7	-	-	PUNCT
cana-5407	184	8	gan	gan	PROPN
cana-5407	184	9	is	be	AUX
cana-5407	184	10	designed	design	VERB
cana-5407	184	11	to	to	PART
cana-5407	184	12	play	play	VERB
cana-5407	184	13	.	.	PUNCT
cana-5407	185	1	hifi	hifi	PROPN
cana-5407	185	2	-	-	PUNCT
cana-5407	185	3	gan	gan	PROPN
cana-5407	185	4	is	be	AUX
cana-5407	185	5	a	a	DET
cana-5407	185	6	generative	generative	ADJ
cana-5407	185	7	adversarial	adversarial	ADJ
cana-5407	185	8	network	network	NOUN
cana-5407	185	9	specifically	specifically	ADV
cana-5407	185	10	developed	develop	VERB
cana-5407	185	11	to	to	PART
cana-5407	185	12	synthesize	synthesize	VERB
cana-5407	185	13	high	high	ADJ
cana-5407	185	14	-	-	PUNCT
cana-5407	185	15	quality	quality	NOUN
cana-5407	185	16	audio	audio	NOUN
cana-5407	185	17	from	from	ADP
cana-5407	185	18	mel	mel	PROPN
cana-5407	185	19	spectrograms	spectrograms	PROPN
cana-5407	185	20	.	.	PUNCT
cana-5407	186	1	it	it	PRON
cana-5407	186	2	synthesizes	synthesize	VERB
cana-5407	186	3	audio	audio	ADJ
cana-5407	186	4	signals	signal	NOUN
cana-5407	186	5	with	with	ADP
cana-5407	186	6	a	a	DET
cana-5407	186	7	high	high	ADJ
cana-5407	186	8	level	level	NOUN
cana-5407	186	9	of	of	ADP
cana-5407	186	10	understanding	understanding	NOUN
cana-5407	186	11	,	,	PUNCT
cana-5407	186	12	with	with	ADP
cana-5407	186	13	smoother	smooth	ADJ
cana-5407	186	14	speech	speech	NOUN
cana-5407	186	15	and	and	CCONJ
cana-5407	186	16	fewer	few	ADJ
cana-5407	186	17	robotic	robotic	ADJ
cana-5407	186	18	-	-	PUNCT
cana-5407	186	19	sounding	sound	VERB
cana-5407	186	20	audio	audio	ADJ
cana-5407	186	21	bits	bit	NOUN
cana-5407	186	22	.	.	PUNCT
cana-5407	187	1	its	its	PRON
cana-5407	187	2	high	high	ADJ
cana-5407	187	3	fidelity	fidelity	NOUN
cana-5407	187	4	without	without	ADP
cana-5407	187	5	the	the	DET
cana-5407	187	6	grainy	grainy	ADJ
cana-5407	187	7	sound	sound	NOUN
cana-5407	187	8	experienced	experience	VERB
cana-5407	187	9	with	with	ADP
cana-5407	187	10	other	other	ADJ
cana-5407	187	11	tts	tts	PROPN
cana-5407	187	12	is	be	AUX
cana-5407	187	13	as	as	ADP
cana-5407	187	14	a	a	DET
cana-5407	187	15	result	result	NOUN
cana-5407	187	16	of	of	ADP
cana-5407	187	17	the	the	DET
cana-5407	187	18	existing	exist	VERB
cana-5407	187	19	speech	speech	NOUN
cana-5407	187	20	dataset	dataset	VERB
cana-5407	187	21	from	from	ADP
cana-5407	187	22	which	which	PRON
cana-5407	187	23	it	it	PRON
cana-5407	187	24	had	have	AUX
cana-5407	187	25	been	be	AUX
cana-5407	187	26	trained	train	VERB
cana-5407	187	27	to	to	PART
cana-5407	187	28	learn	learn	VERB
cana-5407	187	29	the	the	DET
cana-5407	187	30	desired	desire	VERB
cana-5407	187	31	audio	audio	NOUN
cana-5407	187	32	.	.	PUNCT
cana-5407	188	1	to	to	PART
cana-5407	188	2	top	top	VERB
cana-5407	188	3	up	up	ADP
cana-5407	188	4	the	the	DET
cana-5407	188	5	quality	quality	NOUN
cana-5407	188	6	of	of	ADP
cana-5407	188	7	the	the	DET
cana-5407	188	8	audio	audio	NOUN
cana-5407	188	9	,	,	PUNCT
cana-5407	188	10	there	there	PRON
cana-5407	188	11	exist	exist	VERB
cana-5407	188	12	super	super	ADJ
cana-5407	188	13	-	-	ADJ
cana-5407	188	14	resolution	resolution	ADJ
cana-5407	188	15	techniques	technique	NOUN
cana-5407	188	16	.	.	PUNCT
cana-5407	189	1	audio	audio	ADJ
cana-5407	189	2	super	super	ADJ
cana-5407	189	3	-	-	NOUN
cana-5407	189	4	resolution	resolution	NOUN
cana-5407	189	5	is	be	AUX
cana-5407	189	6	an	an	DET
cana-5407	189	7	accomplished	accomplished	ADJ
cana-5407	189	8	computation	computation	NOUN
cana-5407	189	9	technique	technique	NOUN
cana-5407	189	10	that	that	PRON
cana-5407	189	11	involves	involve	VERB
cana-5407	189	12	increasing	increase	VERB
cana-5407	189	13	the	the	DET
cana-5407	189	14	number	number	NOUN
cana-5407	189	15	of	of	ADP
cana-5407	189	16	times	time	NOUN
cana-5407	189	17	the	the	DET
cana-5407	189	18	audio	audio	NOUN
cana-5407	189	19	’s	’s	PART
cana-5407	189	20	analog	analog	NOUN
cana-5407	189	21	to	to	ADP
cana-5407	189	22	digital	digital	ADJ
cana-5407	189	23	sampling	sampling	NOUN
cana-5407	189	24	was	be	AUX
cana-5407	189	25	undertaken	undertake	VERB
cana-5407	189	26	or	or	CCONJ
cana-5407	189	27	improving	improve	VERB
cana-5407	189	28	the	the	DET
cana-5407	189	29	level	level	NOUN
cana-5407	189	30	of	of	ADP
cana-5407	189	31	detail	detail	NOUN
cana-5407	189	32	in	in	ADP
cana-5407	189	33	the	the	DET
cana-5407	189	34	audio	audio	NOUN
cana-5407	189	35	produced	produce	VERB
cana-5407	189	36	.	.	PUNCT
cana-5407	190	1	in	in	ADP
cana-5407	190	2	the	the	DET
cana-5407	190	3	instance	instance	NOUN
cana-5407	190	4	of	of	ADP
cana-5407	190	5	tts	tts	PROPN
cana-5407	190	6	,	,	PUNCT
cana-5407	190	7	super	super	ADJ
cana-5407	190	8	-	-	NOUN
cana-5407	190	9	resolution	resolution	NOUN
cana-5407	190	10	can	can	AUX
cana-5407	190	11	be	be	AUX
cana-5407	190	12	applied	apply	VERB
cana-5407	190	13	where	where	SCONJ
cana-5407	190	14	some	some	DET
cana-5407	190	15	sounds	sound	NOUN
cana-5407	190	16	might	might	AUX
cana-5407	190	17	be	be	AUX
cana-5407	190	18	finer	fine	ADJ
cana-5407	190	19	than	than	SCONJ
cana-5407	190	20	desired	desire	VERB
cana-5407	190	21	,	,	PUNCT
cana-5407	190	22	resulting	result	VERB
cana-5407	190	23	in	in	ADP
cana-5407	190	24	a	a	DET
cana-5407	190	25	sound	sound	NOUN
cana-5407	190	26	that	that	PRON
cana-5407	190	27	may	may	AUX
cana-5407	190	28	not	not	PART
cana-5407	190	29	be	be	AUX
cana-5407	190	30	clear	clear	ADJ
cana-5407	190	31	.	.	PUNCT
cana-5407	191	1	the	the	DET
cana-5407	191	2	super	super	NOUN
cana-5407	191	3	-	-	NOUN
cana-5407	191	4	resolution	resolution	NOUN
cana-5407	191	5	of	of	ADP
cana-5407	191	6	sounds	sound	NOUN
cana-5407	191	7	improves	improve	VERB
cana-5407	191	8	their	their	PRON
cana-5407	191	9	quality	quality	NOUN
cana-5407	191	10	experience	experience	NOUN
cana-5407	191	11	.	.	PUNCT
cana-5407	192	1	communications	communication	NOUN
cana-5407	192	2	on	on	ADP
cana-5407	192	3	applied	apply	VERB
cana-5407	192	4	nonlinear	nonlinear	ADJ
cana-5407	192	5	analysis	analysis	NOUN
cana-5407	192	6	issn	issn	NOUN
cana-5407	192	7	:	:	PUNCT
cana-5407	192	8	1074	1074	NUM
cana-5407	192	9	-	-	PUNCT
cana-5407	192	10	133x	133x	NUM
cana-5407	192	11	vol	vol	VERB
cana-5407	192	12	32	32	NUM
cana-5407	192	13	no	no	NOUN
cana-5407	192	14	.	.	PUNCT
cana-5407	193	1	10s	10	NOUN
cana-5407	193	2	(	(	PUNCT
cana-5407	193	3	2025	2025	NUM
cana-5407	193	4	)	)	PUNCT
cana-5407	193	5	2180	2180	NUM
cana-5407	193	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-5407	193	7	at	at	ADP
cana-5407	193	8	the	the	DET
cana-5407	193	9	top	top	NOUN
cana-5407	193	10	of	of	ADP
cana-5407	193	11	the	the	DET
cana-5407	193	12	hierarchy	hierarchy	NOUN
cana-5407	193	13	are	be	AUX
cana-5407	193	14	the	the	DET
cana-5407	193	15	main	main	ADJ
cana-5407	193	16	processes	process	NOUN
cana-5407	193	17	,	,	PUNCT
cana-5407	193	18	including	include	VERB
cana-5407	193	19	audio	audio	NOUN
cana-5407	193	20	processing	processing	NOUN
cana-5407	193	21	,	,	PUNCT
cana-5407	193	22	where	where	SCONJ
cana-5407	193	23	there	there	PRON
cana-5407	193	24	are	be	VERB
cana-5407	193	25	techniques	technique	NOUN
cana-5407	193	26	such	such	ADJ
cana-5407	193	27	as	as	ADP
cana-5407	193	28	griffin	griffin	NOUN
cana-5407	193	29	-	-	PUNCT
cana-5407	193	30	lim	lim	PROPN
cana-5407	193	31	synthesis	synthesis	NOUN
cana-5407	193	32	,	,	PUNCT
cana-5407	193	33	which	which	PRON
cana-5407	193	34	is	be	AUX
cana-5407	193	35	one	one	NUM
cana-5407	193	36	of	of	ADP
cana-5407	193	37	the	the	DET
cana-5407	193	38	approaches	approach	NOUN
cana-5407	193	39	that	that	PRON
cana-5407	193	40	convert	convert	VERB
cana-5407	193	41	a	a	DET
cana-5407	193	42	mel	mel	PROPN
cana-5407	193	43	spectrogram	spectrogram	NOUN
cana-5407	193	44	back	back	ADV
cana-5407	193	45	into	into	ADP
cana-5407	193	46	an	an	DET
cana-5407	193	47	audio	audio	ADJ
cana-5407	193	48	waveform	waveform	NOUN
cana-5407	193	49	.	.	PUNCT
cana-5407	194	1	griffin	griffin	PROPN
cana-5407	194	2	-	-	PUNCT
cana-5407	194	3	lim	lim	PROPN
cana-5407	194	4	synthesis	synthesis	NOUN
cana-5407	194	5	tends	tend	VERB
cana-5407	194	6	to	to	PART
cana-5407	194	7	generate	generate	VERB
cana-5407	194	8	a	a	DET
cana-5407	194	9	certain	certain	ADJ
cana-5407	194	10	amount	amount	NOUN
cana-5407	194	11	of	of	ADP
cana-5407	194	12	artifacts	artifact	NOUN
cana-5407	194	13	while	while	SCONJ
cana-5407	194	14	it	it	PRON
cana-5407	194	15	can	can	AUX
cana-5407	194	16	improve	improve	VERB
cana-5407	194	17	the	the	DET
cana-5407	194	18	generation	generation	NOUN
cana-5407	194	19	of	of	ADP
cana-5407	194	20	audio	audio	NOUN
cana-5407	194	21	from	from	ADP
cana-5407	194	22	spectrograms	spectrogram	NOUN
cana-5407	194	23	in	in	ADP
cana-5407	194	24	conjunction	conjunction	NOUN
cana-5407	194	25	with	with	ADP
cana-5407	194	26	models	model	NOUN
cana-5407	194	27	such	such	ADJ
cana-5407	194	28	as	as	ADP
cana-5407	194	29	hifi	hifi	NOUN
cana-5407	194	30	-	-	PUNCT
cana-5407	194	31	gan	gin	VERB
cana-5407	194	32	by	by	ADP
cana-5407	194	33	iteratively	iteratively	ADV
cana-5407	194	34	approximating	approximate	VERB
cana-5407	194	35	spectrograms	spectrogram	NOUN
cana-5407	194	36	.	.	PUNCT
cana-5407	195	1	moreover	moreover	ADV
cana-5407	195	2	,	,	PUNCT
cana-5407	195	3	audio	audio	ADJ
cana-5407	195	4	signals	signal	NOUN
cana-5407	195	5	go	go	VERB
cana-5407	195	6	through	through	ADP
cana-5407	195	7	filters	filter	NOUN
cana-5407	195	8	and	and	CCONJ
cana-5407	195	9	normalization	normalization	NOUN
cana-5407	195	10	to	to	PART
cana-5407	195	11	constrain	constrain	VERB
cana-5407	195	12	the	the	DET
cana-5407	195	13	volume	volume	NOUN
cana-5407	195	14	and	and	CCONJ
cana-5407	195	15	pattern	pattern	NOUN
cana-5407	195	16	of	of	ADP
cana-5407	195	17	the	the	DET
cana-5407	195	18	sound	sound	NOUN
cana-5407	195	19	in	in	ADP
cana-5407	195	20	terms	term	NOUN
cana-5407	195	21	of	of	ADP
cana-5407	195	22	achieving	achieve	VERB
cana-5407	195	23	the	the	DET
cana-5407	195	24	necessary	necessary	ADJ
cana-5407	195	25	audio	audio	NOUN
cana-5407	195	26	based	base	VERB
cana-5407	195	27	on	on	ADP
cana-5407	195	28	the	the	DET
cana-5407	195	29	predetermined	predetermine	VERB
cana-5407	195	30	allowable	allowable	ADJ
cana-5407	195	31	audio	audio	ADJ
cana-5407	195	32	standards	standard	NOUN
cana-5407	195	33	.	.	PUNCT
cana-5407	196	1	results	result	NOUN
cana-5407	196	2	:	:	PUNCT
cana-5407	196	3	this	this	DET
cana-5407	196	4	section	section	NOUN
cana-5407	196	5	reveals	reveal	VERB
cana-5407	196	6	the	the	DET
cana-5407	196	7	results	result	NOUN
cana-5407	196	8	of	of	ADP
cana-5407	196	9	simulations	simulation	NOUN
cana-5407	196	10	performed	perform	VERB
cana-5407	196	11	with	with	ADP
cana-5407	196	12	the	the	DET
cana-5407	196	13	recalibrated	recalibrate	VERB
cana-5407	196	14	voice	voice	NOUN
cana-5407	196	15	conversion	conversion	NOUN
cana-5407	196	16	model	model	NOUN
cana-5407	196	17	above	above	ADV
cana-5407	196	18	,	,	PUNCT
cana-5407	196	19	responsible	responsible	ADJ
cana-5407	196	20	for	for	ADP
cana-5407	196	21	upgrading	upgrade	VERB
cana-5407	196	22	the	the	DET
cana-5407	196	23	quality	quality	NOUN
cana-5407	196	24	of	of	ADP
cana-5407	196	25	speech	speech	NOUN
cana-5407	196	26	transformation	transformation	NOUN
cana-5407	196	27	,	,	PUNCT
cana-5407	196	28	with	with	ADP
cana-5407	196	29	a	a	DET
cana-5407	196	30	specific	specific	ADJ
cana-5407	196	31	focus	focus	NOUN
cana-5407	196	32	on	on	ADP
cana-5407	196	33	prosody	prosody	NOUN
cana-5407	196	34	,	,	PUNCT
cana-5407	196	35	naturalness	naturalness	NOUN
cana-5407	196	36	,	,	PUNCT
cana-5407	196	37	and	and	CCONJ
cana-5407	196	38	intonation	intonation	NOUN
cana-5407	196	39	on	on	ADP
cana-5407	196	40	generated	generate	VERB
cana-5407	196	41	speech	speech	NOUN
cana-5407	196	42	.	.	PUNCT
cana-5407	197	1	the	the	DET
cana-5407	197	2	experiments	experiment	NOUN
cana-5407	197	3	implied	imply	VERB
cana-5407	197	4	the	the	DET
cana-5407	197	5	collection	collection	NOUN
cana-5407	197	6	of	of	ADP
cana-5407	197	7	a	a	DET
cana-5407	197	8	dataset	dataset	NOUN
cana-5407	197	9	made	make	VERB
cana-5407	197	10	up	up	ADP
cana-5407	197	11	of	of	ADP
cana-5407	197	12	1000	1000	NUM
cana-5407	197	13	spoken	spoken	ADJ
cana-5407	197	14	english	english	ADJ
cana-5407	197	15	audio	audio	NOUN
cana-5407	197	16	recordings	recording	NOUN
cana-5407	197	17	.	.	PUNCT
cana-5407	198	1	the	the	DET
cana-5407	198	2	audio	audio	ADJ
cana-5407	198	3	data	datum	NOUN
cana-5407	198	4	was	be	AUX
cana-5407	198	5	equally	equally	ADV
cana-5407	198	6	spread	spread	VERB
cana-5407	198	7	among	among	ADP
cana-5407	198	8	fifty	fifty	NUM
cana-5407	198	9	male	male	ADJ
cana-5407	198	10	and	and	CCONJ
cana-5407	198	11	female	female	ADJ
cana-5407	198	12	english	english	ADJ
cana-5407	198	13	speakers	speaker	NOUN
cana-5407	198	14	with	with	ADP
cana-5407	198	15	varying	vary	VERB
cana-5407	198	16	accents	accent	NOUN
cana-5407	198	17	.	.	PUNCT
cana-5407	199	1	the	the	DET
cana-5407	199	2	model	model	NOUN
cana-5407	199	3	was	be	AUX
cana-5407	199	4	trained	train	VERB
cana-5407	199	5	within	within	ADP
cana-5407	199	6	a	a	DET
cana-5407	199	7	tesla	tesla	NOUN
cana-5407	199	8	v100	v100	PROPN
cana-5407	199	9	gpu	gpu	NOUN
cana-5407	199	10	throughout	throughout	ADP
cana-5407	199	11	250	250	NUM
cana-5407	199	12	epochs	epoch	NOUN
cana-5407	199	13	,	,	PUNCT
cana-5407	199	14	with	with	ADP
cana-5407	199	15	batch	batch	NOUN
cana-5407	199	16	size	size	NOUN
cana-5407	199	17	as	as	ADP
cana-5407	199	18	three	three	NUM
cana-5407	199	19	.	.	PUNCT
cana-5407	200	1	the	the	DET
cana-5407	200	2	quality	quality	NOUN
cana-5407	200	3	was	be	AUX
cana-5407	200	4	assessed	assess	VERB
cana-5407	200	5	using	use	VERB
cana-5407	200	6	the	the	DET
cana-5407	200	7	mel	mel	NOUN
cana-5407	200	8	-	-	ADJ
cana-5407	200	9	cepstral	cepstral	ADJ
cana-5407	200	10	distortion	distortion	NOUN
cana-5407	200	11	by	by	ADP
cana-5407	200	12	means	mean	VERB
cana-5407	200	13	perception	perception	NOUN
cana-5407	200	14	tests	test	NOUN
cana-5407	200	15	online	online	ADV
cana-5407	200	16	.	.	PUNCT
cana-5407	201	1	the	the	DET
cana-5407	201	2	first	first	ADJ
cana-5407	201	3	component	component	NOUN
cana-5407	201	4	of	of	ADP
cana-5407	201	5	the	the	DET
cana-5407	201	6	process	process	NOUN
cana-5407	201	7	is	be	AUX
cana-5407	201	8	data	data	NOUN
cana-5407	201	9	collection	collection	NOUN
cana-5407	201	10	,	,	PUNCT
cana-5407	201	11	in	in	ADP
cana-5407	201	12	which	which	PRON
cana-5407	201	13	a	a	DET
cana-5407	201	14	diverse	diverse	ADJ
cana-5407	201	15	range	range	NOUN
cana-5407	201	16	of	of	ADP
cana-5407	201	17	speech	speech	NOUN
cana-5407	201	18	samples	sample	NOUN
cana-5407	201	19	are	be	AUX
cana-5407	201	20	required	require	VERB
cana-5407	201	21	across	across	ADP
cana-5407	201	22	genders	gender	NOUN
cana-5407	201	23	,	,	PUNCT
cana-5407	201	24	accents	accent	NOUN
cana-5407	201	25	,	,	PUNCT
cana-5407	201	26	and	and	CCONJ
cana-5407	201	27	languages	language	NOUN
cana-5407	201	28	among	among	ADP
cana-5407	201	29	others	other	NOUN
cana-5407	201	30	.	.	PUNCT
cana-5407	202	1	from	from	ADP
cana-5407	202	2	the	the	DET
cana-5407	202	3	collection	collection	NOUN
cana-5407	202	4	,	,	PUNCT
cana-5407	202	5	the	the	DET
cana-5407	202	6	audio	audio	NOUN
cana-5407	202	7	speech	speech	NOUN
cana-5407	202	8	is	be	AUX
cana-5407	202	9	transcribed	transcribe	VERB
cana-5407	202	10	to	to	PART
cana-5407	202	11	text	text	VERB
cana-5407	202	12	while	while	SCONJ
cana-5407	202	13	concurrently	concurrently	ADV
cana-5407	202	14	segmenting	segment	VERB
cana-5407	202	15	the	the	DET
cana-5407	202	16	audio	audio	ADJ
cana-5407	202	17	files	file	NOUN
cana-5407	202	18	so	so	SCONJ
cana-5407	202	19	that	that	SCONJ
cana-5407	202	20	each	each	PRON
cana-5407	202	21	of	of	ADP
cana-5407	202	22	the	the	DET
cana-5407	202	23	segments	segment	NOUN
cana-5407	202	24	aligns	align	VERB
cana-5407	202	25	with	with	ADP
cana-5407	202	26	one	one	NUM
cana-5407	202	27	textual	textual	ADJ
cana-5407	202	28	equivalent	equivalent	NOUN
cana-5407	202	29	.	.	PUNCT
cana-5407	203	1	concretely	concretely	ADV
cana-5407	203	2	,	,	PUNCT
cana-5407	203	3	the	the	DET
cana-5407	203	4	target	target	NOUN
cana-5407	203	5	audio	audio	NOUN
cana-5407	203	6	needs	need	NOUN
cana-5407	203	7	to	to	PART
cana-5407	203	8	be	be	AUX
cana-5407	203	9	segmented	segment	VERB
cana-5407	203	10	into	into	ADP
cana-5407	203	11	a	a	DET
cana-5407	203	12	set	set	NOUN
cana-5407	203	13	of	of	ADP
cana-5407	203	14	well	well	ADV
cana-5407	203	15	-	-	PUNCT
cana-5407	203	16	defined	define	VERB
cana-5407	203	17	units	unit	NOUN
cana-5407	203	18	one	one	NUM
cana-5407	203	19	of	of	ADP
cana-5407	203	20	them	they	PRON
cana-5407	203	21	is	be	AUX
cana-5407	203	22	sentences	sentence	NOUN
cana-5407	203	23	or	or	CCONJ
cana-5407	203	24	phrases	phrase	NOUN
cana-5407	203	25	”	"	PUNCT
cana-5407	203	26	.	.	PUNCT
cana-5407	204	1	the	the	DET
cana-5407	204	2	keyed	keyed	ADJ
cana-5407	204	3	transcriptions	transcription	NOUN
cana-5407	204	4	are	be	AUX
cana-5407	204	5	then	then	ADV
cana-5407	204	6	converted	convert	VERB
cana-5407	204	7	into	into	ADP
cana-5407	204	8	the	the	DET
cana-5407	204	9	most	most	ADV
cana-5407	204	10	appropriate	appropriate	ADJ
cana-5407	204	11	format	format	NOUN
cana-5407	204	12	,	,	PUNCT
cana-5407	204	13	which	which	PRON
cana-5407	204	14	in	in	ADP
cana-5407	204	15	this	this	DET
cana-5407	204	16	case	case	NOUN
cana-5407	204	17	will	will	AUX
cana-5407	204	18	be	be	AUX
cana-5407	204	19	mel	mel	NOUN
cana-5407	204	20	-	-	PUNCT
cana-5407	204	21	spectrograms	spectrogram	NOUN
cana-5407	204	22	.	.	PUNCT
cana-5407	205	1	these	these	DET
cana-5407	205	2	representations	representation	NOUN
cana-5407	205	3	offer	offer	VERB
cana-5407	205	4	a	a	DET
cana-5407	205	5	visual	visual	ADJ
cana-5407	205	6	inspection	inspection	NOUN
cana-5407	205	7	of	of	ADP
cana-5407	205	8	the	the	DET
cana-5407	205	9	power	power	NOUN
cana-5407	205	10	spectrum	spectrum	NOUN
cana-5407	205	11	of	of	ADP
cana-5407	205	12	sound	sound	NOUN
cana-5407	205	13	and	and	CCONJ
cana-5407	205	14	time	time	NOUN
cana-5407	205	15	,	,	PUNCT
cana-5407	205	16	manifesting	manifest	VERB
cana-5407	205	17	the	the	DET
cana-5407	205	18	most	most	ADV
cana-5407	205	19	suited	suited	ADJ
cana-5407	205	20	feature	feature	NOUN
cana-5407	205	21	set	set	VERB
cana-5407	205	22	for	for	ADP
cana-5407	205	23	speech	speech	NOUN
cana-5407	205	24	synthesis	synthesis	NOUN
cana-5407	205	25	.	.	PUNCT
cana-5407	206	1	the	the	DET
cana-5407	206	2	spectrum	spectrum	NOUN
cana-5407	206	3	of	of	ADP
cana-5407	206	4	sample	sample	NOUN
cana-5407	206	5	voices	voice	NOUN
cana-5407	206	6	in	in	ADP
cana-5407	206	7	the	the	DET
cana-5407	206	8	dataset	dataset	NOUN
cana-5407	206	9	are	be	AUX
cana-5407	206	10	depicted	depict	VERB
cana-5407	206	11	in	in	ADP
cana-5407	206	12	figure	figure	NOUN
cana-5407	206	13	2	2	NUM
cana-5407	206	14	.	.	PUNCT
cana-5407	206	15	figure	figure	NOUN
cana-5407	206	16	2(a	2(a	NUM
cana-5407	206	17	)	)	PUNCT
cana-5407	206	18	represents	represent	VERB
cana-5407	206	19	the	the	DET
cana-5407	206	20	spectrum	spectrum	NOUN
cana-5407	206	21	of	of	ADP
cana-5407	206	22	male	male	ADJ
cana-5407	206	23	speech	speech	NOUN
cana-5407	206	24	,	,	PUNCT
cana-5407	206	25	and	and	CCONJ
cana-5407	206	26	its	its	PRON
cana-5407	206	27	mel	mel	PROPN
cana-5407	206	28	frequency	frequency	NOUN
cana-5407	206	29	cepstral	cepstral	ADJ
cana-5407	206	30	coefficients	coefficient	NOUN
cana-5407	206	31	(	(	PUNCT
cana-5407	206	32	mfcc	mfcc	NOUN
cana-5407	206	33	)	)	PUNCT
cana-5407	206	34	plot	plot	NOUN
cana-5407	206	35	is	be	AUX
cana-5407	206	36	shown	show	VERB
cana-5407	206	37	in	in	ADP
cana-5407	206	38	figure	figure	NOUN
cana-5407	206	39	2(b	2(b	NUM
cana-5407	206	40	)	)	PUNCT
cana-5407	206	41	.	.	PUNCT
cana-5407	207	1	similarly	similarly	ADV
cana-5407	207	2	,	,	PUNCT
cana-5407	207	3	figure	figure	NOUN
cana-5407	207	4	3(a	3(a	NUM
cana-5407	207	5	)	)	PUNCT
cana-5407	207	6	and	and	CCONJ
cana-5407	207	7	3(b	3(b	NUM
cana-5407	207	8	)	)	PUNCT
cana-5407	207	9	represents	represent	VERB
cana-5407	207	10	the	the	DET
cana-5407	207	11	spectrum	spectrum	NOUN
cana-5407	207	12	and	and	CCONJ
cana-5407	207	13	mfcc	mfcc	NOUN
cana-5407	207	14	plot	plot	NOUN
cana-5407	207	15	of	of	ADP
cana-5407	207	16	female	female	ADJ
cana-5407	207	17	speech	speech	NOUN
cana-5407	207	18	.	.	PUNCT
cana-5407	208	1	mfcc	mfcc	NOUN
cana-5407	208	2	are	be	AUX
cana-5407	208	3	widely	widely	ADV
cana-5407	208	4	used	use	VERB
cana-5407	208	5	in	in	ADP
cana-5407	208	6	the	the	DET
cana-5407	208	7	processing	processing	NOUN
cana-5407	208	8	of	of	ADP
cana-5407	208	9	speech	speech	NOUN
cana-5407	208	10	and	and	CCONJ
cana-5407	208	11	audio	audio	NOUN
cana-5407	208	12	and	and	CCONJ
cana-5407	208	13	used	use	VERB
cana-5407	208	14	for	for	ADP
cana-5407	208	15	such	such	ADJ
cana-5407	208	16	tasks	task	NOUN
cana-5407	208	17	as	as	ADP
cana-5407	208	18	speech	speech	NOUN
cana-5407	208	19	recognition	recognition	NOUN
cana-5407	208	20	and	and	CCONJ
cana-5407	208	21	music	music	NOUN
cana-5407	208	22	analysis	analysis	NOUN
cana-5407	208	23	.	.	PUNCT
cana-5407	209	1	in	in	ADP
cana-5407	209	2	an	an	DET
cana-5407	209	3	mfcc	mfcc	NOUN
cana-5407	209	4	plot	plot	NOUN
cana-5407	209	5	,	,	PUNCT
cana-5407	209	6	the	the	DET
cana-5407	209	7	coefficients	coefficient	NOUN
cana-5407	209	8	are	be	AUX
cana-5407	209	9	plotted	plot	VERB
cana-5407	209	10	against	against	ADP
cana-5407	209	11	time	time	NOUN
cana-5407	209	12	,	,	PUNCT
cana-5407	209	13	showing	show	VERB
cana-5407	209	14	how	how	SCONJ
cana-5407	209	15	the	the	DET
cana-5407	209	16	sound	sound	NOUN
cana-5407	209	17	’s	’s	PART
cana-5407	209	18	spectral	spectral	ADJ
cana-5407	209	19	properties	property	NOUN
cana-5407	209	20	change	change	VERB
cana-5407	209	21	over	over	ADP
cana-5407	209	22	time	time	NOUN
cana-5407	209	23	.	.	PUNCT
cana-5407	210	1	the	the	DET
cana-5407	210	2	first	first	ADJ
cana-5407	210	3	axis	axis	NOUN
cana-5407	210	4	on	on	ADP
cana-5407	210	5	this	this	DET
cana-5407	210	6	plot	plot	NOUN
cana-5407	210	7	is	be	AUX
cana-5407	210	8	a	a	DET
cana-5407	210	9	function	function	NOUN
cana-5407	210	10	of	of	ADP
cana-5407	210	11	time	time	NOUN
cana-5407	210	12	,	,	PUNCT
cana-5407	210	13	while	while	SCONJ
cana-5407	210	14	the	the	DET
cana-5407	210	15	second	second	NOUN
cana-5407	210	16	is	be	AUX
cana-5407	210	17	the	the	DET
cana-5407	210	18	values	value	NOUN
cana-5407	210	19	of	of	ADP
cana-5407	210	20	the	the	DET
cana-5407	210	21	individual	individual	ADJ
cana-5407	210	22	coefficients	coefficient	NOUN
cana-5407	210	23	.	.	PUNCT
cana-5407	211	1	the	the	DET
cana-5407	211	2	lowest	low	ADJ
cana-5407	211	3	coefficient	coefficient	NOUN
cana-5407	211	4	is	be	AUX
cana-5407	211	5	typically	typically	ADV
cana-5407	211	6	located	locate	VERB
cana-5407	211	7	at	at	ADP
cana-5407	211	8	the	the	DET
cana-5407	211	9	bottom	bottom	NOUN
cana-5407	211	10	,	,	PUNCT
cana-5407	211	11	while	while	SCONJ
cana-5407	211	12	increasing	increase	VERB
cana-5407	211	13	coefficients	coefficient	NOUN
cana-5407	211	14	are	be	AUX
cana-5407	211	15	on	on	ADP
cana-5407	211	16	top	top	NOUN
cana-5407	211	17	.	.	PUNCT
cana-5407	212	1	a	a	DET
cana-5407	212	2	few	few	ADJ
cana-5407	212	3	other	other	ADJ
cana-5407	212	4	things	thing	NOUN
cana-5407	212	5	to	to	PART
cana-5407	212	6	consider	consider	VERB
cana-5407	212	7	are	be	AUX
cana-5407	212	8	as	as	SCONJ
cana-5407	212	9	follows	follow	VERB
cana-5407	212	10	:	:	PUNCT
cana-5407	212	11	the	the	DET
cana-5407	212	12	x	x	X
cana-5407	212	13	-	-	ADJ
cana-5407	212	14	axis	axis	NOUN
cana-5407	212	15	is	be	AUX
cana-5407	212	16	the	the	DET
cana-5407	212	17	time	time	NOUN
cana-5407	212	18	axis	axis	NOUN
cana-5407	212	19	:	:	PUNCT
cana-5407	212	20	this	this	PRON
cana-5407	212	21	represents	represent	VERB
cana-5407	212	22	the	the	DET
cana-5407	212	23	movement	movement	NOUN
cana-5407	212	24	of	of	ADP
cana-5407	212	25	time	time	NOUN
cana-5407	212	26	as	as	SCONJ
cana-5407	212	27	the	the	DET
cana-5407	212	28	sound	sound	NOUN
cana-5407	212	29	signal	signal	NOUN
cana-5407	212	30	is	be	AUX
cana-5407	212	31	analyzed	analyze	VERB
cana-5407	212	32	.	.	PUNCT
cana-5407	213	1	the	the	DET
cana-5407	213	2	coefficient	coefficient	PROPN
cana-5407	213	3	index	index	NOUN
cana-5407	213	4	axis	axis	NOUN
cana-5407	213	5	refers	refer	VERB
cana-5407	213	6	to	to	ADP
cana-5407	213	7	the	the	DET
cana-5407	213	8	y	y	NOUN
cana-5407	213	9	-	-	PUNCT
cana-5407	213	10	axis	axis	NOUN
cana-5407	213	11	:	:	PUNCT
cana-5407	213	12	for	for	ADP
cana-5407	213	13	this	this	DET
cana-5407	213	14	axis	axis	NOUN
cana-5407	213	15	,	,	PUNCT
cana-5407	213	16	the	the	DET
cana-5407	213	17	indices	index	NOUN
cana-5407	213	18	of	of	ADP
cana-5407	213	19	the	the	DET
cana-5407	213	20	cepstral	cepstral	ADJ
cana-5407	213	21	coefficients	coefficient	NOUN
cana-5407	213	22	are	be	AUX
cana-5407	213	23	often	often	ADV
cana-5407	213	24	displayed	display	VERB
cana-5407	213	25	.	.	PUNCT
cana-5407	214	1	color	color	NOUN
cana-5407	214	2	or	or	CCONJ
cana-5407	214	3	intensity	intensity	NOUN
cana-5407	214	4	:	:	PUNCT
cana-5407	214	5	the	the	DET
cana-5407	214	6	third	third	ADJ
cana-5407	214	7	property	property	NOUN
cana-5407	214	8	often	often	ADV
cana-5407	214	9	displayed	display	VERB
cana-5407	214	10	is	be	AUX
cana-5407	214	11	the	the	DET
cana-5407	214	12	color	color	NOUN
cana-5407	214	13	or	or	CCONJ
cana-5407	214	14	intensity	intensity	NOUN
cana-5407	214	15	of	of	ADP
cana-5407	214	16	the	the	DET
cana-5407	214	17	value	value	NOUN
cana-5407	214	18	of	of	ADP
cana-5407	214	19	the	the	DET
cana-5407	214	20	mfccs	mfccs	ADJ
cana-5407	214	21	summaries	summary	NOUN
cana-5407	214	22	.	.	PUNCT
cana-5407	215	1	(	(	PUNCT
cana-5407	215	2	a	a	X
cana-5407	215	3	)	)	PUNCT
cana-5407	215	4	spectrum	spectrum	NOUN
cana-5407	215	5	of	of	ADP
cana-5407	215	6	speech	speech	NOUN
cana-5407	215	7	signal-1	signal-1	NUM
cana-5407	215	8	communications	communication	NOUN
cana-5407	215	9	on	on	ADP
cana-5407	215	10	applied	apply	VERB
cana-5407	215	11	nonlinear	nonlinear	ADJ
cana-5407	215	12	analysis	analysis	NOUN
cana-5407	215	13	issn	issn	NOUN
cana-5407	215	14	:	:	PUNCT
cana-5407	215	15	1074	1074	NUM
cana-5407	215	16	-	-	PUNCT
cana-5407	215	17	133x	133x	NUM
cana-5407	215	18	vol	vol	VERB
cana-5407	215	19	32	32	NUM
cana-5407	215	20	no	no	NOUN
cana-5407	215	21	.	.	PUNCT
cana-5407	216	1	10s	10	NOUN
cana-5407	216	2	(	(	PUNCT
cana-5407	216	3	2025	2025	NUM
cana-5407	216	4	)	)	PUNCT
cana-5407	216	5	2181	2181	NUM
cana-5407	217	1	https://internationalpubls.com	https://internationalpubls.com	X
cana-5407	217	2	(	(	PUNCT
cana-5407	217	3	b	b	NOUN
cana-5407	217	4	)	)	PUNCT
cana-5407	217	5	mfcc	mfcc	NOUN
cana-5407	217	6	of	of	ADP
cana-5407	217	7	the	the	DET
cana-5407	217	8	speech	speech	NOUN
cana-5407	217	9	signal	signal	NOUN
cana-5407	217	10	-1	-1	PUNCT
cana-5407	217	11	.	.	PUNCT
cana-5407	218	1	figure	figure	NOUN
cana-5407	218	2	2	2	NUM
cana-5407	218	3	:	:	PUNCT
cana-5407	218	4	male	male	ADJ
cana-5407	218	5	speech	speech	NOUN
cana-5407	218	6	text	text	NOUN
cana-5407	218	7	:	:	PUNCT
cana-5407	218	8	“	"	PUNCT
cana-5407	218	9	the	the	DET
cana-5407	218	10	fusion	fusion	NOUN
cana-5407	218	11	of	of	ADP
cana-5407	218	12	jaz	jaz	PROPN
cana-5407	218	13	and	and	CCONJ
cana-5407	218	14	classical	classical	ADJ
cana-5407	218	15	music	music	NOUN
cana-5407	218	16	joneris	joneris	NOUN
cana-5407	218	17	creates	create	VERB
cana-5407	218	18	a	a	DET
cana-5407	218	19	unique	unique	ADJ
cana-5407	218	20	sound	sound	NOUN
cana-5407	218	21	”	"	PUNCT
cana-5407	218	22	(	(	PUNCT
cana-5407	218	23	a	a	PRON
cana-5407	218	24	)	)	PUNCT
cana-5407	218	25	spectrum	spectrum	NOUN
cana-5407	218	26	of	of	ADP
cana-5407	218	27	speech	speech	NOUN
cana-5407	218	28	signal-2	signal-2	PROPN
cana-5407	218	29	.	.	PUNCT
cana-5407	219	1	(	(	PUNCT
cana-5407	219	2	b	b	NOUN
cana-5407	219	3	)	)	PUNCT
cana-5407	219	4	mfcc	mfcc	NOUN
cana-5407	219	5	of	of	ADP
cana-5407	219	6	speech	speech	NOUN
cana-5407	219	7	signal-2	signal-2	NUM
cana-5407	219	8	figure	figure	NOUN
cana-5407	219	9	3	3	NUM
cana-5407	219	10	:	:	PUNCT
cana-5407	219	11	female	female	ADJ
cana-5407	219	12	speech	speech	NOUN
cana-5407	219	13	text	text	NOUN
cana-5407	219	14	:	:	PUNCT
cana-5407	219	15	thunderous	thunderous	ADJ
cana-5407	219	16	applause	applause	NOUN
cana-5407	219	17	erupted	erupt	VERB
cana-5407	219	18	as	as	SCONJ
cana-5407	219	19	the	the	DET
cana-5407	219	20	curtain	curtain	NOUN
cana-5407	219	21	fell	fall	VERB
cana-5407	219	22	marking	mark	VERB
cana-5407	219	23	the	the	DET
cana-5407	219	24	end	end	NOUN
cana-5407	219	25	of	of	ADP
cana-5407	219	26	the	the	DET
cana-5407	219	27	spectacular	spectacular	ADJ
cana-5407	219	28	performance	performance	NOUN
cana-5407	219	29	it	it	PRON
cana-5407	219	30	is	be	AUX
cana-5407	219	31	used	use	VERB
cana-5407	219	32	to	to	PART
cana-5407	219	33	visualize	visualize	VERB
cana-5407	219	34	the	the	DET
cana-5407	219	35	dimension	dimension	NOUN
cana-5407	219	36	of	of	ADP
cana-5407	219	37	an	an	DET
cana-5407	219	38	onset	onset	NOUN
cana-5407	219	39	per	per	ADP
cana-5407	219	40	time	time	NOUN
cana-5407	219	41	frame	frame	NOUN
cana-5407	219	42	.	.	PUNCT
cana-5407	220	1	it	it	PRON
cana-5407	220	2	includes	include	VERB
cana-5407	220	3	color	color	NOUN
cana-5407	220	4	-	-	PUNCT
cana-5407	220	5	coded	code	VERB
cana-5407	220	6	autoscale	autoscale	NOUN
cana-5407	220	7	and	and	CCONJ
cana-5407	220	8	optional	optional	ADJ
cana-5407	220	9	horizontal	horizontal	ADJ
cana-5407	220	10	bar	bar	NOUN
cana-5407	220	11	.	.	PUNCT
cana-5407	221	1	the	the	DET
cana-5407	221	2	scale	scale	NOUN
cana-5407	221	3	is	be	AUX
cana-5407	221	4	adjusted	adjust	VERB
cana-5407	221	5	to	to	PART
cana-5407	221	6	match	match	VERB
cana-5407	221	7	the	the	DET
cana-5407	221	8	sequence	sequence	NOUN
cana-5407	221	9	and	and	CCONJ
cana-5407	221	10	is	be	AUX
cana-5407	221	11	displayed	display	VERB
cana-5407	221	12	with	with	ADP
cana-5407	221	13	the	the	DET
cana-5407	221	14	data	datum	NOUN
cana-5407	221	15	.	.	PUNCT
cana-5407	222	1	after	after	SCONJ
cana-5407	222	2	the	the	DET
cana-5407	222	3	data	data	NOUN
cana-5407	222	4	is	be	AUX
cana-5407	222	5	prepared	prepare	VERB
cana-5407	222	6	,	,	PUNCT
cana-5407	222	7	training	training	NOUN
cana-5407	222	8	of	of	ADP
cana-5407	222	9	the	the	DET
cana-5407	222	10	proposed	propose	VERB
cana-5407	222	11	tacotron	tacotron	NOUN
cana-5407	222	12	2	2	NUM
cana-5407	222	13	model	model	NOUN
cana-5407	222	14	starts	start	VERB
cana-5407	222	15	.	.	PUNCT
cana-5407	223	1	the	the	DET
cana-5407	223	2	model	model	NOUN
cana-5407	223	3	architecture	architecture	NOUN
cana-5407	223	4	is	be	AUX
cana-5407	223	5	usually	usually	ADV
cana-5407	223	6	divided	divide	VERB
cana-5407	223	7	into	into	ADP
cana-5407	223	8	two	two	NUM
cana-5407	223	9	distinct	distinct	ADJ
cana-5407	223	10	sections	section	NOUN
cana-5407	223	11	:	:	PUNCT
cana-5407	223	12	the	the	DET
cana-5407	223	13	encoder	encoder	NOUN
cana-5407	223	14	and	and	CCONJ
cana-5407	223	15	the	the	DET
cana-5407	223	16	decoder	decoder	NOUN
cana-5407	223	17	.	.	PUNCT
cana-5407	224	1	an	an	DET
cana-5407	224	2	encoder	encoder	NOUN
cana-5407	224	3	is	be	AUX
cana-5407	224	4	trained	train	VERB
cana-5407	224	5	on	on	ADP
cana-5407	224	6	input	input	NOUN
cana-5407	224	7	text	text	NOUN
cana-5407	224	8	to	to	PART
cana-5407	224	9	convert	convert	VERB
cana-5407	224	10	characters	character	NOUN
cana-5407	224	11	into	into	ADP
cana-5407	224	12	an	an	DET
cana-5407	224	13	isotropous	isotropous	ADJ
cana-5407	224	14	,	,	PUNCT
cana-5407	224	15	high	high	ADJ
cana-5407	224	16	-	-	PUNCT
cana-5407	224	17	dimensional	dimensional	ADJ
cana-5407	224	18	embedding	embedding	NOUN
cana-5407	224	19	.	.	PUNCT
cana-5407	225	1	this	this	DET
cana-5407	225	2	representation	representation	NOUN
cana-5407	225	3	captures	capture	VERB
cana-5407	225	4	the	the	DET
cana-5407	225	5	nuanced	nuanced	ADJ
cana-5407	225	6	language	language	NOUN
cana-5407	225	7	structure	structure	NOUN
cana-5407	225	8	and	and	CCONJ
cana-5407	225	9	context	context	NOUN
cana-5407	225	10	required	require	VERB
cana-5407	225	11	to	to	PART
cana-5407	225	12	produce	produce	VERB
cana-5407	225	13	speech	speech	NOUN
cana-5407	225	14	.	.	PUNCT
cana-5407	226	1	on	on	ADP
cana-5407	226	2	the	the	DET
cana-5407	226	3	other	other	ADJ
cana-5407	226	4	hand	hand	NOUN
cana-5407	226	5	,	,	PUNCT
cana-5407	226	6	the	the	DET
cana-5407	226	7	decoder	decoder	NOUN
cana-5407	226	8	is	be	AUX
cana-5407	226	9	responsible	responsible	ADJ
cana-5407	226	10	for	for	ADP
cana-5407	226	11	converting	convert	VERB
cana-5407	226	12	this	this	PRON
cana-5407	226	13	into	into	ADP
cana-5407	226	14	audio	audio	ADJ
cana-5407	226	15	data	datum	NOUN
cana-5407	226	16	or	or	CCONJ
cana-5407	226	17	output	output	NOUN
cana-5407	226	18	.	.	PUNCT
cana-5407	227	1	it	it	PRON
cana-5407	227	2	predicts	predict	VERB
cana-5407	227	3	the	the	DET
cana-5407	227	4	melcommunications	melcommunication	NOUN
cana-5407	227	5	on	on	ADP
cana-5407	227	6	applied	apply	VERB
cana-5407	227	7	nonlinear	nonlinear	ADJ
cana-5407	227	8	analysis	analysis	NOUN
cana-5407	227	9	issn	issn	NOUN
cana-5407	227	10	:	:	PUNCT
cana-5407	227	11	1074	1074	NUM
cana-5407	227	12	-	-	PUNCT
cana-5407	227	13	133x	133x	NUM
cana-5407	227	14	vol	vol	VERB
cana-5407	227	15	32	32	NUM
cana-5407	227	16	no	no	NOUN
cana-5407	227	17	.	.	PUNCT
cana-5407	228	1	10s	10	NOUN
cana-5407	228	2	(	(	PUNCT
cana-5407	228	3	2025	2025	NUM
cana-5407	228	4	)	)	PUNCT
cana-5407	228	5	2182	2182	NUM
cana-5407	228	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-5407	228	7	spectrogram	spectrogram	NOUN
cana-5407	228	8	frame	frame	NOUN
cana-5407	228	9	by	by	ADP
cana-5407	228	10	frame	frame	NOUN
cana-5407	228	11	,	,	PUNCT
cana-5407	228	12	the	the	DET
cana-5407	228	13	graphical	graphical	ADJ
cana-5407	228	14	form	form	NOUN
cana-5407	228	15	of	of	ADP
cana-5407	228	16	audio	audio	NOUN
cana-5407	228	17	’s	’s	PART
cana-5407	228	18	spectral	spectral	ADJ
cana-5407	228	19	properties	property	NOUN
cana-5407	228	20	.	.	PUNCT
cana-5407	229	1	more	more	ADV
cana-5407	229	2	precisely	precisely	ADV
cana-5407	229	3	,	,	PUNCT
cana-5407	229	4	training	training	NOUN
cana-5407	229	5	involves	involve	VERB
cana-5407	229	6	learning	learn	VERB
cana-5407	229	7	the	the	DET
cana-5407	229	8	model	model	NOUN
cana-5407	229	9	’s	’s	PART
cana-5407	229	10	parameters	parameter	NOUN
cana-5407	229	11	so	so	SCONJ
cana-5407	229	12	that	that	SCONJ
cana-5407	229	13	the	the	DET
cana-5407	229	14	mel	mel	PROPN
cana-5407	229	15	-	-	PUNCT
cana-5407	229	16	spectrogram	spectrogram	NOUN
cana-5407	229	17	it	it	PRON
cana-5407	229	18	outputs	output	VERB
cana-5407	229	19	,	,	PUNCT
cana-5407	229	20	when	when	SCONJ
cana-5407	229	21	passed	pass	VERB
cana-5407	229	22	to	to	ADP
cana-5407	229	23	a	a	DET
cana-5407	229	24	proper	proper	ADJ
cana-5407	229	25	decoder	decoder	NOUN
cana-5407	229	26	,	,	PUNCT
cana-5407	229	27	resembles	resemble	VERB
cana-5407	229	28	the	the	DET
cana-5407	229	29	ground	ground	NOUN
cana-5407	229	30	truth	truth	NOUN
cana-5407	229	31	mel	mel	PROPN
cana-5407	229	32	-	-	PUNCT
cana-5407	229	33	spectrogram	spectrogram	NOUN
cana-5407	229	34	as	as	ADV
cana-5407	229	35	closely	closely	ADV
cana-5407	229	36	as	as	ADP
cana-5407	229	37	possible	possible	ADJ
cana-5407	229	38	,	,	PUNCT
cana-5407	229	39	and	and	CCONJ
cana-5407	229	40	is	be	AUX
cana-5407	229	41	done	do	VERB
cana-5407	229	42	by	by	ADP
cana-5407	229	43	optimization	optimization	NOUN
cana-5407	229	44	using	use	VERB
cana-5407	229	45	a	a	DET
cana-5407	229	46	specially	specially	ADV
cana-5407	229	47	-	-	PUNCT
cana-5407	229	48	designed	design	VERB
cana-5407	229	49	loss	loss	NOUN
cana-5407	229	50	function	function	NOUN
cana-5407	229	51	.	.	PUNCT
cana-5407	230	1	(	(	PUNCT
cana-5407	230	2	a	a	X
cana-5407	230	3	)	)	PUNCT
cana-5407	230	4	training	training	NOUN
cana-5407	230	5	at	at	ADP
cana-5407	230	6	epoch	epoch	PROPN
cana-5407	230	7	-0	-0	PUNCT
cana-5407	230	8	validation	validation	NOUN
cana-5407	230	9	loss	loss	NOUN
cana-5407	230	10	:	:	PUNCT
cana-5407	230	11	1.007171	1.007171	NUM
cana-5407	230	12	(	(	PUNCT
cana-5407	230	13	b	b	NOUN
cana-5407	230	14	)	)	PUNCT
cana-5407	230	15	training	training	NOUN
cana-5407	230	16	at	at	ADP
cana-5407	230	17	epoch-22	epoch-22	PROPN
cana-5407	230	18	validation	validation	NOUN
cana-5407	230	19	loss	loss	NOUN
cana-5407	230	20	:	:	PUNCT
cana-5407	230	21	0.212975	0.212975	NUM
cana-5407	230	22	(	(	PUNCT
cana-5407	230	23	c	c	NOUN
cana-5407	230	24	)	)	PUNCT
cana-5407	230	25	training	training	NOUN
cana-5407	230	26	at	at	ADP
cana-5407	230	27	epoch	epoch	NOUN
cana-5407	230	28	-45	-45	PUNCT
cana-5407	230	29	validation	validation	NOUN
cana-5407	230	30	loss	loss	NOUN
cana-5407	230	31	:	:	PUNCT
cana-5407	230	32	0.212975	0.212975	NUM
cana-5407	230	33	figure	figure	NOUN
cana-5407	230	34	4	4	NUM
cana-5407	230	35	:	:	PUNCT
cana-5407	230	36	encoder	encoder	NOUN
cana-5407	230	37	vs	vs	ADP
cana-5407	230	38	decoder	decoder	NOUN
cana-5407	230	39	time	time	NOUN
cana-5407	230	40	step	step	NOUN
cana-5407	230	41	plot	plot	NOUN
cana-5407	230	42	at	at	ADP
cana-5407	230	43	different	different	ADJ
cana-5407	230	44	epochs	epoch	NOUN
cana-5407	230	45	figure	figure	VERB
cana-5407	230	46	5	5	NUM
cana-5407	230	47	represents	represent	VERB
cana-5407	230	48	a	a	DET
cana-5407	230	49	heat	heat	NOUN
cana-5407	230	50	map	map	NOUN
cana-5407	230	51	representing	represent	VERB
cana-5407	230	52	the	the	DET
cana-5407	230	53	dependence	dependence	NOUN
cana-5407	230	54	of	of	ADP
cana-5407	230	55	encoder	encoder	NOUN
cana-5407	230	56	timesteps	timestep	VERB
cana-5407	230	57	on	on	ADP
cana-5407	230	58	decoder	decoder	NOUN
cana-5407	230	59	timesteps	timestep	NOUN
cana-5407	230	60	in	in	ADP
cana-5407	230	61	a	a	DET
cana-5407	230	62	sequence	sequence	NOUN
cana-5407	230	63	-	-	PUNCT
cana-5407	230	64	to	to	ADP
cana-5407	230	65	-	-	PUNCT
cana-5407	230	66	sequence	sequence	NOUN
cana-5407	230	67	model	model	NOUN
cana-5407	230	68	,	,	PUNCT
cana-5407	230	69	which	which	PRON
cana-5407	230	70	is	be	AUX
cana-5407	230	71	typically	typically	ADV
cana-5407	230	72	applied	apply	VERB
cana-5407	230	73	in	in	ADP
cana-5407	230	74	machine	machine	NOUN
cana-5407	230	75	learning	learning	NOUN
cana-5407	230	76	for	for	ADP
cana-5407	230	77	language	language	NOUN
cana-5407	230	78	translation	translation	NOUN
cana-5407	230	79	,	,	PUNCT
cana-5407	230	80	text	text	NOUN
cana-5407	230	81	summarization	summarization	NOUN
cana-5407	230	82	or	or	CCONJ
cana-5407	230	83	speech	speech	NOUN
cana-5407	230	84	recognition	recognition	NOUN
cana-5407	230	85	.	.	PUNCT
cana-5407	231	1	more	more	ADV
cana-5407	231	2	specifically	specifically	ADV
cana-5407	231	3	,	,	PUNCT
cana-5407	231	4	an	an	DET
cana-5407	231	5	encoder	encoder	NOUN
cana-5407	231	6	reads	read	VERB
cana-5407	231	7	the	the	DET
cana-5407	231	8	input	input	NOUN
cana-5407	231	9	data	datum	NOUN
cana-5407	231	10	,	,	PUNCT
cana-5407	231	11	for	for	ADP
cana-5407	231	12	example	example	NOUN
cana-5407	231	13	,	,	PUNCT
cana-5407	231	14	a	a	DET
cana-5407	231	15	sentence	sentence	NOUN
cana-5407	231	16	in	in	ADP
cana-5407	231	17	the	the	DET
cana-5407	231	18	source	source	NOUN
cana-5407	231	19	language	language	NOUN
cana-5407	231	20	,	,	PUNCT
cana-5407	231	21	gradually	gradually	ADV
cana-5407	231	22	,	,	PUNCT
cana-5407	231	23	one	one	NUM
cana-5407	231	24	timestep	timestep	NOUN
cana-5407	231	25	after	after	ADP
cana-5407	231	26	another	another	PRON
cana-5407	231	27	,	,	PUNCT
cana-5407	231	28	so	so	SCONJ
cana-5407	231	29	as	as	SCONJ
cana-5407	231	30	to	to	PART
cana-5407	231	31	create	create	VERB
cana-5407	231	32	a	a	DET
cana-5407	231	33	representation	representation	NOUN
cana-5407	231	34	of	of	ADP
cana-5407	231	35	it	it	PRON
cana-5407	231	36	.	.	PUNCT
cana-5407	232	1	after	after	ADP
cana-5407	232	2	that	that	PRON
cana-5407	232	3	,	,	PUNCT
cana-5407	232	4	the	the	DET
cana-5407	232	5	decoder	decoder	NOUN
cana-5407	232	6	produces	produce	VERB
cana-5407	232	7	the	the	DET
cana-5407	232	8	output	output	NOUN
cana-5407	232	9	data	datum	NOUN
cana-5407	232	10	sequentially	sequentially	ADV
cana-5407	232	11	,	,	PUNCT
cana-5407	232	12	timestep	timestep	NOUN
cana-5407	232	13	by	by	ADP
cana-5407	232	14	timestep	timestep	NOUN
cana-5407	232	15	,	,	PUNCT
cana-5407	232	16	and	and	CCONJ
cana-5407	232	17	this	this	DET
cana-5407	232	18	process	process	NOUN
cana-5407	232	19	may	may	AUX
cana-5407	232	20	depend	depend	VERB
cana-5407	232	21	on	on	ADP
cana-5407	232	22	the	the	DET
cana-5407	232	23	model	model	NOUN
cana-5407	232	24	of	of	ADP
cana-5407	232	25	the	the	DET
cana-5407	232	26	encoder	encoder	NOUN
cana-5407	232	27	.	.	PUNCT
cana-5407	233	1	in	in	ADP
cana-5407	233	2	figure	figure	NOUN
cana-5407	233	3	5	5	NUM
cana-5407	233	4	,	,	PUNCT
cana-5407	233	5	the	the	DET
cana-5407	233	6	communications	communication	NOUN
cana-5407	233	7	on	on	ADP
cana-5407	233	8	applied	apply	VERB
cana-5407	233	9	nonlinear	nonlinear	ADJ
cana-5407	233	10	analysis	analysis	NOUN
cana-5407	233	11	issn	issn	NOUN
cana-5407	233	12	:	:	PUNCT
cana-5407	233	13	1074	1074	NUM
cana-5407	233	14	-	-	PUNCT
cana-5407	233	15	133x	133x	NUM
cana-5407	233	16	vol	vol	VERB
cana-5407	233	17	32	32	NUM
cana-5407	233	18	no	no	NOUN
cana-5407	233	19	.	.	PUNCT
cana-5407	234	1	10s	10	NOUN
cana-5407	234	2	(	(	PUNCT
cana-5407	234	3	2025	2025	NUM
cana-5407	234	4	)	)	PUNCT
cana-5407	234	5	2183	2183	NUM
cana-5407	235	1	https://internationalpubls.com	https://internationalpubls.com	X
cana-5407	235	2	presented	present	VERB
cana-5407	235	3	heat	heat	NOUN
cana-5407	235	4	map	map	NOUN
cana-5407	235	5	is	be	AUX
cana-5407	235	6	a	a	DET
cana-5407	235	7	visualization	visualization	NOUN
cana-5407	235	8	of	of	ADP
cana-5407	235	9	the	the	DET
cana-5407	235	10	attention	attention	NOUN
cana-5407	235	11	weights	weight	NOUN
cana-5407	235	12	or	or	CCONJ
cana-5407	235	13	alignments	alignment	NOUN
cana-5407	235	14	between	between	ADP
cana-5407	235	15	each	each	DET
cana-5407	235	16	encoder	encoder	NOUN
cana-5407	235	17	timestep	timestep	NOUN
cana-5407	235	18	on	on	ADP
cana-5407	235	19	the	the	DET
cana-5407	235	20	y	y	NOUN
cana-5407	235	21	-	-	PUNCT
cana-5407	235	22	axis	axis	NOUN
cana-5407	235	23	and	and	CCONJ
cana-5407	235	24	each	each	DET
cana-5407	235	25	decoder	decoder	NOUN
cana-5407	235	26	timestep	timestep	NOUN
cana-5407	235	27	on	on	ADP
cana-5407	235	28	the	the	DET
cana-5407	235	29	x	x	NOUN
cana-5407	235	30	-	-	NOUN
cana-5407	235	31	axis	axis	NOUN
cana-5407	235	32	.	.	PUNCT
cana-5407	236	1	the	the	DET
cana-5407	236	2	color	color	NOUN
cana-5407	236	3	scale	scale	NOUN
cana-5407	236	4	on	on	ADP
cana-5407	236	5	the	the	DET
cana-5407	236	6	right	right	NOUN
cana-5407	236	7	shows	show	VERB
cana-5407	236	8	the	the	DET
cana-5407	236	9	strength	strength	NOUN
cana-5407	236	10	of	of	ADP
cana-5407	236	11	the	the	DET
cana-5407	236	12	attention	attention	NOUN
cana-5407	236	13	or	or	CCONJ
cana-5407	236	14	alignment	alignment	NOUN
cana-5407	236	15	,	,	PUNCT
cana-5407	236	16	and	and	CCONJ
cana-5407	236	17	warmer	warm	ADJ
cana-5407	236	18	colors	color	NOUN
cana-5407	236	19	indicate	indicate	VERB
cana-5407	236	20	higher	high	ADJ
cana-5407	236	21	weights	weight	NOUN
cana-5407	236	22	in	in	ADP
cana-5407	236	23	the	the	DET
cana-5407	236	24	attention	attention	NOUN
cana-5407	236	25	vector	vector	NOUN
cana-5407	236	26	while	while	SCONJ
cana-5407	236	27	cooler	cool	ADJ
cana-5407	236	28	colors	color	NOUN
cana-5407	236	29	indicate	indicate	VERB
cana-5407	236	30	lower	low	ADJ
cana-5407	236	31	weights	weight	NOUN
cana-5407	236	32	.	.	PUNCT
cana-5407	237	1	the	the	DET
cana-5407	237	2	diagonal	diagonal	ADJ
cana-5407	237	3	pattern	pattern	NOUN
cana-5407	237	4	seen	see	VERB
cana-5407	237	5	in	in	ADP
cana-5407	237	6	the	the	DET
cana-5407	237	7	figure	figure	NOUN
cana-5407	237	8	indicates	indicate	VERB
cana-5407	237	9	that	that	SCONJ
cana-5407	237	10	each	each	DET
cana-5407	237	11	decoder	decoder	NOUN
cana-5407	237	12	timestep	timestep	NOUN
cana-5407	237	13	seems	seem	VERB
cana-5407	237	14	to	to	PART
cana-5407	237	15	pay	pay	VERB
cana-5407	237	16	significant	significant	ADJ
cana-5407	237	17	attention	attention	NOUN
cana-5407	237	18	mainly	mainly	ADV
cana-5407	237	19	to	to	ADP
cana-5407	237	20	the	the	DET
cana-5407	237	21	corresponding	correspond	VERB
cana-5407	237	22	encoder	encoder	NOUN
cana-5407	237	23	timestep	timestep	NOUN
cana-5407	237	24	or	or	CCONJ
cana-5407	237	25	encoder	encoder	NOUN
cana-5407	237	26	timesteps	timestep	VERB
cana-5407	237	27	close	close	ADV
cana-5407	237	28	by	by	ADV
cana-5407	237	29	in	in	ADP
cana-5407	237	30	the	the	DET
cana-5407	237	31	input	input	NOUN
cana-5407	237	32	sequence	sequence	NOUN
cana-5407	237	33	.	.	PUNCT
cana-5407	238	1	this	this	DET
cana-5407	238	2	pattern	pattern	NOUN
cana-5407	238	3	is	be	AUX
cana-5407	238	4	evidence	evidence	NOUN
cana-5407	238	5	of	of	ADP
cana-5407	238	6	a	a	DET
cana-5407	238	7	monotonic	monotonic	ADJ
cana-5407	238	8	alignment	alignment	NOUN
cana-5407	238	9	between	between	ADP
cana-5407	238	10	the	the	DET
cana-5407	238	11	input	input	NOUN
cana-5407	238	12	sequence	sequence	NOUN
cana-5407	238	13	and	and	CCONJ
cana-5407	238	14	the	the	DET
cana-5407	238	15	output	output	NOUN
cana-5407	238	16	sequence	sequence	NOUN
cana-5407	238	17	.	.	PUNCT
cana-5407	239	1	it	it	PRON
cana-5407	239	2	is	be	AUX
cana-5407	239	3	most	most	ADV
cana-5407	239	4	common	common	ADJ
cana-5407	239	5	in	in	ADP
cana-5407	239	6	topics	topic	NOUN
cana-5407	239	7	such	such	ADJ
cana-5407	239	8	as	as	ADP
cana-5407	239	9	machine	machine	NOUN
cana-5407	239	10	translation	translation	NOUN
cana-5407	239	11	where	where	SCONJ
cana-5407	239	12	the	the	DET
cana-5407	239	13	input	input	NOUN
cana-5407	239	14	and	and	CCONJ
cana-5407	239	15	output	output	NOUN
cana-5407	239	16	sequence	sequence	NOUN
cana-5407	239	17	’s	’s	PART
cana-5407	239	18	individual	individual	ADJ
cana-5407	239	19	sentences	sentence	NOUN
cana-5407	239	20	have	have	VERB
cana-5407	239	21	a	a	DET
cana-5407	239	22	comparable	comparable	ADJ
cana-5407	239	23	structure	structure	NOUN
cana-5407	239	24	.	.	PUNCT
cana-5407	240	1	after	after	ADP
cana-5407	240	2	training	train	VERB
cana-5407	240	3	the	the	DET
cana-5407	240	4	speech	speech	NOUN
cana-5407	240	5	synthesis	synthesis	NOUN
cana-5407	240	6	model	model	NOUN
cana-5407	240	7	based	base	VERB
cana-5407	240	8	on	on	ADP
cana-5407	240	9	proposed	propose	VERB
cana-5407	240	10	tacotron	tacotron	NOUN
cana-5407	240	11	2	2	NUM
cana-5407	240	12	,	,	PUNCT
cana-5407	240	13	one	one	PRON
cana-5407	240	14	also	also	ADV
cana-5407	240	15	needs	need	VERB
cana-5407	240	16	to	to	PART
cana-5407	240	17	train	train	VERB
cana-5407	240	18	its	its	PRON
cana-5407	240	19	voice	voice	NOUN
cana-5407	240	20	conversion	conversion	NOUN
cana-5407	240	21	model	model	NOUN
cana-5407	240	22	.	.	PUNCT
cana-5407	241	1	this	this	DET
cana-5407	241	2	model	model	NOUN
cana-5407	241	3	aims	aim	VERB
cana-5407	241	4	to	to	ADP
cana-5407	241	5	fine	fine	ADJ
cana-5407	241	6	-	-	PUNCT
cana-5407	241	7	tune	tune	NOUN
cana-5407	241	8	the	the	DET
cana-5407	241	9	speech	speech	NOUN
cana-5407	241	10	creation	creation	NOUN
cana-5407	241	11	process	process	NOUN
cana-5407	241	12	so	so	SCONJ
cana-5407	241	13	that	that	SCONJ
cana-5407	241	14	it	it	PRON
cana-5407	241	15	resembles	resemble	VERB
cana-5407	241	16	the	the	DET
cana-5407	241	17	voice	voice	NOUN
cana-5407	241	18	of	of	ADP
cana-5407	241	19	the	the	DET
cana-5407	241	20	target	target	NOUN
cana-5407	241	21	one	one	NUM
cana-5407	241	22	more	more	ADV
cana-5407	241	23	closely	closely	ADV
cana-5407	241	24	.	.	PUNCT
cana-5407	242	1	the	the	DET
cana-5407	242	2	voice	voice	NOUN
cana-5407	242	3	of	of	ADP
cana-5407	242	4	the	the	DET
cana-5407	242	5	speech	speech	NOUN
cana-5407	242	6	can	can	AUX
cana-5407	242	7	be	be	AUX
cana-5407	242	8	altered	alter	VERB
cana-5407	242	9	and	and	CCONJ
cana-5407	242	10	include	include	VERB
cana-5407	242	11	changes	change	NOUN
cana-5407	242	12	in	in	ADP
cana-5407	242	13	pitch	pitch	NOUN
cana-5407	242	14	and	and	CCONJ
cana-5407	242	15	timbre	timbre	NOUN
cana-5407	242	16	and	and	CCONJ
cana-5407	242	17	also	also	ADV
cana-5407	242	18	rhythm	rhythm	PROPN
cana-5407	242	19	.	.	PUNCT
cana-5407	243	1	this	this	DET
cana-5407	243	2	process	process	NOUN
cana-5407	243	3	calls	call	VERB
cana-5407	243	4	for	for	ADP
cana-5407	243	5	additional	additional	ADJ
cana-5407	243	6	data	datum	NOUN
cana-5407	243	7	from	from	ADP
cana-5407	243	8	the	the	DET
cana-5407	243	9	target	target	NOUN
cana-5407	243	10	voice	voice	NOUN
cana-5407	243	11	,	,	PUNCT
cana-5407	243	12	and	and	CCONJ
cana-5407	243	13	it	it	PRON
cana-5407	243	14	might	might	AUX
cana-5407	243	15	also	also	ADV
cana-5407	243	16	use	use	VERB
cana-5407	243	17	voice	voice	NOUN
cana-5407	243	18	embedding	embed	VERB
cana-5407	243	19	and	and	CCONJ
cana-5407	243	20	style	style	NOUN
cana-5407	243	21	transfer	transfer	NOUN
cana-5407	243	22	to	to	PART
cana-5407	243	23	learn	learn	VERB
cana-5407	243	24	intensively	intensively	ADV
cana-5407	243	25	the	the	DET
cana-5407	243	26	unique	unique	ADJ
cana-5407	243	27	properties	property	NOUN
cana-5407	243	28	of	of	ADP
cana-5407	243	29	the	the	DET
cana-5407	243	30	target	target	NOUN
cana-5407	243	31	voice	voice	NOUN
cana-5407	243	32	and	and	CCONJ
cana-5407	243	33	imitate	imitate	VERB
cana-5407	243	34	them	they	PRON
cana-5407	243	35	.	.	PUNCT
cana-5407	244	1	the	the	DET
cana-5407	244	2	next	next	ADJ
cana-5407	244	3	step	step	NOUN
cana-5407	244	4	is	be	AUX
cana-5407	244	5	fine	fine	ADV
cana-5407	244	6	-	-	PUNCT
cana-5407	244	7	tuning	tune	VERB
cana-5407	244	8	the	the	DET
cana-5407	244	9	model	model	NOUN
cana-5407	244	10	.	.	PUNCT
cana-5407	245	1	this	this	DET
cana-5407	245	2	procedure	procedure	NOUN
cana-5407	245	3	fine	fine	NOUN
cana-5407	245	4	-	-	PUNCT
cana-5407	245	5	tunes	tune	NOUN
cana-5407	245	6	the	the	DET
cana-5407	245	7	model	model	NOUN
cana-5407	245	8	to	to	PART
cana-5407	245	9	be	be	AUX
cana-5407	245	10	more	more	ADV
cana-5407	245	11	sensitive	sensitive	ADJ
cana-5407	245	12	to	to	ADP
cana-5407	245	13	the	the	DET
cana-5407	245	14	subtleties	subtlety	NOUN
cana-5407	245	15	of	of	ADP
cana-5407	245	16	target	target	NOUN
cana-5407	245	17	voice	voice	NOUN
cana-5407	245	18	characteristics	characteristic	NOUN
cana-5407	245	19	,	,	PUNCT
cana-5407	245	20	which	which	PRON
cana-5407	245	21	guarantees	guarantee	VERB
cana-5407	245	22	that	that	SCONJ
cana-5407	245	23	while	while	SCONJ
cana-5407	245	24	the	the	DET
cana-5407	245	25	synthesized	synthesize	VERB
cana-5407	245	26	voice	voice	NOUN
cana-5407	245	27	is	be	AUX
cana-5407	245	28	made	make	VERB
cana-5407	245	29	to	to	PART
cana-5407	245	30	sound	sound	VERB
cana-5407	245	31	natural	natural	ADJ
cana-5407	245	32	,	,	PUNCT
cana-5407	245	33	it	it	PRON
cana-5407	245	34	also	also	ADV
cana-5407	245	35	sounds	sound	VERB
cana-5407	245	36	as	as	ADV
cana-5407	245	37	close	close	ADJ
cana-5407	245	38	as	as	ADP
cana-5407	245	39	possible	possible	ADJ
cana-5407	245	40	to	to	ADP
cana-5407	245	41	the	the	DET
cana-5407	245	42	intended	intend	VERB
cana-5407	245	43	target	target	NOUN
cana-5407	245	44	output	output	NOUN
cana-5407	245	45	voice	voice	NOUN
cana-5407	245	46	.	.	PUNCT
cana-5407	246	1	additionally	additionally	ADV
cana-5407	246	2	,	,	PUNCT
cana-5407	246	3	optimization	optimization	NOUN
cana-5407	246	4	can	can	AUX
cana-5407	246	5	involve	involve	VERB
cana-5407	246	6	some	some	DET
cana-5407	246	7	model	model	NOUN
cana-5407	246	8	refinement	refinement	NOUN
cana-5407	246	9	;	;	PUNCT
cana-5407	246	10	for	for	ADP
cana-5407	246	11	example	example	NOUN
cana-5407	246	12	,	,	PUNCT
cana-5407	246	13	the	the	DET
cana-5407	246	14	model	model	NOUN
cana-5407	246	15	can	can	AUX
cana-5407	246	16	be	be	AUX
cana-5407	246	17	modified	modify	VERB
cana-5407	246	18	to	to	PART
cana-5407	246	19	generate	generate	VERB
cana-5407	246	20	voices	voice	NOUN
cana-5407	246	21	faster	fast	ADV
cana-5407	246	22	or	or	CCONJ
cana-5407	246	23	produce	produce	VERB
cana-5407	246	24	more	more	ADJ
cana-5407	246	25	articulate	articulate	ADJ
cana-5407	246	26	synthesized	synthesize	VERB
cana-5407	246	27	speech	speech	NOUN
cana-5407	246	28	.	.	PUNCT
cana-5407	247	1	(	(	PUNCT
cana-5407	247	2	a	a	X
cana-5407	247	3	)	)	PUNCT
cana-5407	247	4	heatmap	heatmap	NOUN
cana-5407	247	5	and	and	CCONJ
cana-5407	247	6	encoder	encoder	NOUN
cana-5407	247	7	vs	vs	ADP
cana-5407	247	8	decoder	decoder	NOUN
cana-5407	247	9	time	time	NOUN
cana-5407	247	10	step	step	NOUN
cana-5407	247	11	plot	plot	NOUN
cana-5407	247	12	of	of	ADP
cana-5407	247	13	output	output	NOUN
cana-5407	247	14	wave	wave	NOUN
cana-5407	247	15	file	file	NOUN
cana-5407	247	16	(	(	PUNCT
cana-5407	247	17	b	b	NOUN
cana-5407	247	18	)	)	PUNCT
cana-5407	247	19	spectrum	spectrum	NOUN
cana-5407	247	20	of	of	ADP
cana-5407	247	21	output	output	NOUN
cana-5407	247	22	wave	wave	NOUN
cana-5407	247	23	file	file	NOUN
cana-5407	247	24	figure	figure	NOUN
cana-5407	247	25	5	5	NUM
cana-5407	247	26	:	:	PUNCT
cana-5407	247	27	visual	visual	ADJ
cana-5407	247	28	representation	representation	NOUN
cana-5407	247	29	of	of	ADP
cana-5407	247	30	converted	convert	VERB
cana-5407	247	31	speech	speech	NOUN
cana-5407	247	32	wave	wave	NOUN
cana-5407	247	33	file	file	NOUN
cana-5407	247	34	communications	communication	NOUN
cana-5407	247	35	on	on	ADP
cana-5407	247	36	applied	apply	VERB
cana-5407	247	37	nonlinear	nonlinear	ADJ
cana-5407	247	38	analysis	analysis	NOUN
cana-5407	247	39	issn	issn	NOUN
cana-5407	247	40	:	:	PUNCT
cana-5407	247	41	1074	1074	NUM
cana-5407	247	42	-	-	PUNCT
cana-5407	247	43	133x	133x	NUM
cana-5407	247	44	vol	vol	VERB
cana-5407	247	45	32	32	NUM
cana-5407	247	46	no	no	NOUN
cana-5407	247	47	.	.	PUNCT
cana-5407	248	1	10s	10	NOUN
cana-5407	248	2	(	(	PUNCT
cana-5407	248	3	2025	2025	NUM
cana-5407	248	4	)	)	PUNCT
cana-5407	248	5	2184	2184	NUM
cana-5407	249	1	https://internationalpubls.com	https://internationalpubls.com	X
cana-5407	249	2	statistical	statistical	ADJ
cana-5407	249	3	analysis	analysis	NOUN
cana-5407	249	4	was	be	AUX
cana-5407	249	5	performed	perform	VERB
cana-5407	249	6	for	for	ADP
cana-5407	249	7	results	result	NOUN
cana-5407	249	8	validation	validation	NOUN
cana-5407	249	9	.	.	PUNCT
cana-5407	250	1	in	in	ADP
cana-5407	250	2	summary	summary	NOUN
cana-5407	250	3	,	,	PUNCT
cana-5407	250	4	the	the	DET
cana-5407	250	5	proposed	propose	VERB
cana-5407	250	6	novel	novel	NOUN
cana-5407	250	7	tacotron	tacotron	PROPN
cana-5407	250	8	2	2	NUM
cana-5407	250	9	model	model	NOUN
cana-5407	250	10	,	,	PUNCT
cana-5407	250	11	in	in	ADP
cana-5407	250	12	its	its	PRON
cana-5407	250	13	enhanced	enhanced	ADJ
cana-5407	250	14	form	form	NOUN
cana-5407	250	15	,	,	PUNCT
cana-5407	250	16	yield	yield	VERB
cana-5407	250	17	remarkable	remarkable	ADJ
cana-5407	250	18	enhancements	enhancement	NOUN
cana-5407	250	19	in	in	ADP
cana-5407	250	20	terms	term	NOUN
cana-5407	250	21	of	of	ADP
cana-5407	250	22	speech	speech	NOUN
cana-5407	250	23	quality	quality	NOUN
cana-5407	250	24	and	and	CCONJ
cana-5407	250	25	conversion	conversion	NOUN
cana-5407	250	26	accuracy	accuracy	NOUN
cana-5407	250	27	.	.	PUNCT
cana-5407	251	1	these	these	DET
cana-5407	251	2	enhancements	enhancement	NOUN
cana-5407	251	3	results	result	VERB
cana-5407	251	4	from	from	ADP
cana-5407	251	5	employing	employ	VERB
cana-5407	251	6	novel	novel	ADJ
cana-5407	251	7	model	model	NOUN
cana-5407	251	8	architectures	architecture	NOUN
cana-5407	251	9	with	with	ADP
cana-5407	251	10	prosody	prosody	NOUN
cana-5407	251	11	prediction	prediction	NOUN
cana-5407	251	12	and	and	CCONJ
cana-5407	251	13	voice	voice	NOUN
cana-5407	251	14	modulation	modulation	NOUN
cana-5407	251	15	.	.	PUNCT
cana-5407	252	1	these	these	DET
cana-5407	252	2	research	research	NOUN
cana-5407	252	3	results	result	NOUN
cana-5407	252	4	,	,	PUNCT
cana-5407	252	5	thus	thus	ADV
cana-5407	252	6	,	,	PUNCT
cana-5407	252	7	validate	validate	VERB
cana-5407	252	8	the	the	DET
cana-5407	252	9	hypothesis	hypothesis	NOUN
cana-5407	252	10	that	that	PRON
cana-5407	252	11	the	the	DET
cana-5407	252	12	implementation	implementation	NOUN
cana-5407	252	13	of	of	ADP
cana-5407	252	14	enhancements	enhancement	NOUN
cana-5407	252	15	to	to	ADP
cana-5407	252	16	the	the	DET
cana-5407	252	17	tacotron	tacotron	ADJ
cana-5407	252	18	2	2	NUM
cana-5407	252	19	model	model	NOUN
cana-5407	252	20	will	will	AUX
cana-5407	252	21	considerably	considerably	ADV
cana-5407	252	22	enhance	enhance	VERB
cana-5407	252	23	the	the	DET
cana-5407	252	24	naturalness	naturalness	NOUN
cana-5407	252	25	and	and	CCONJ
cana-5407	252	26	conversion	conversion	NOUN
cana-5407	252	27	accuracy	accuracy	NOUN
cana-5407	252	28	.	.	PUNCT
cana-5407	253	1	hence	hence	ADV
cana-5407	253	2	,	,	PUNCT
cana-5407	253	3	they	they	PRON
cana-5407	253	4	demonstrate	demonstrate	VERB
cana-5407	253	5	that	that	SCONJ
cana-5407	253	6	advanced	advanced	ADJ
cana-5407	253	7	neural	neural	ADJ
cana-5407	253	8	architectures	architecture	NOUN
cana-5407	253	9	hold	hold	VERB
cana-5407	253	10	tremendous	tremendous	ADJ
cana-5407	253	11	potential	potential	NOUN
cana-5407	253	12	in	in	ADP
cana-5407	253	13	the	the	DET
cana-5407	253	14	domain	domain	NOUN
cana-5407	253	15	of	of	ADP
cana-5407	253	16	speech	speech	NOUN
cana-5407	253	17	synthesis	synthesis	NOUN
cana-5407	253	18	and	and	CCONJ
cana-5407	253	19	enable	enable	VERB
cana-5407	253	20	the	the	DET
cana-5407	253	21	development	development	NOUN
cana-5407	253	22	of	of	ADP
cana-5407	253	23	synthetic	synthetic	ADJ
cana-5407	253	24	speech	speech	NOUN
cana-5407	253	25	for	for	ADP
cana-5407	253	26	highly	highly	ADV
cana-5407	253	27	sophisticated	sophisticated	ADJ
cana-5407	253	28	and	and	CCONJ
cana-5407	253	29	expressive	expressive	ADJ
cana-5407	253	30	models	model	NOUN
cana-5407	253	31	.	.	PUNCT
cana-5407	254	1	conclusions	conclusion	NOUN
cana-5407	254	2	:	:	PUNCT
cana-5407	254	3	the	the	DET
cana-5407	254	4	paper	paper	NOUN
cana-5407	254	5	presents	present	VERB
cana-5407	254	6	a	a	DET
cana-5407	254	7	comprehensive	comprehensive	ADJ
cana-5407	254	8	study	study	NOUN
cana-5407	254	9	on	on	ADP
cana-5407	254	10	the	the	DET
cana-5407	254	11	application	application	NOUN
cana-5407	254	12	of	of	ADP
cana-5407	254	13	deep	deep	ADJ
cana-5407	254	14	learning	learning	NOUN
cana-5407	254	15	techniques	technique	NOUN
cana-5407	254	16	in	in	ADP
cana-5407	254	17	the	the	DET
cana-5407	254	18	field	field	NOUN
cana-5407	254	19	of	of	ADP
cana-5407	254	20	voice	voice	NOUN
cana-5407	254	21	conversion	conversion	NOUN
cana-5407	254	22	,	,	PUNCT
cana-5407	254	23	with	with	ADP
cana-5407	254	24	an	an	DET
cana-5407	254	25	emphasis	emphasis	NOUN
cana-5407	254	26	on	on	ADP
cana-5407	254	27	improving	improve	VERB
cana-5407	254	28	the	the	DET
cana-5407	254	29	authenticity	authenticity	NOUN
cana-5407	254	30	and	and	CCONJ
cana-5407	254	31	functionality	functionality	NOUN
cana-5407	254	32	of	of	ADP
cana-5407	254	33	the	the	DET
cana-5407	254	34	converted	converted	ADJ
cana-5407	254	35	voices	voice	NOUN
cana-5407	254	36	.	.	PUNCT
cana-5407	255	1	our	our	PRON
cana-5407	255	2	proposed	propose	VERB
cana-5407	255	3	model	model	NOUN
cana-5407	255	4	integrates	integrate	VERB
cana-5407	255	5	automatic	automatic	ADJ
cana-5407	255	6	speech	speech	NOUN
cana-5407	255	7	recognition	recognition	NOUN
cana-5407	255	8	with	with	ADP
cana-5407	255	9	a	a	DET
cana-5407	255	10	sophisticated	sophisticated	ADJ
cana-5407	255	11	voice	voice	NOUN
cana-5407	255	12	conversion	conversion	NOUN
cana-5407	255	13	framework	framework	NOUN
cana-5407	255	14	to	to	PART
cana-5407	255	15	address	address	VERB
cana-5407	255	16	the	the	DET
cana-5407	255	17	prevalent	prevalent	ADJ
cana-5407	255	18	challenges	challenge	NOUN
cana-5407	255	19	in	in	ADP
cana-5407	255	20	the	the	DET
cana-5407	255	21	domain	domain	NOUN
cana-5407	255	22	.	.	PUNCT
cana-5407	256	1	the	the	DET
cana-5407	256	2	experimental	experimental	ADJ
cana-5407	256	3	results	result	NOUN
cana-5407	256	4	highlight	highlight	VERB
cana-5407	256	5	the	the	DET
cana-5407	256	6	model	model	NOUN
cana-5407	256	7	's	's	PART
cana-5407	256	8	efficacy	efficacy	NOUN
cana-5407	256	9	in	in	ADP
cana-5407	256	10	producing	produce	VERB
cana-5407	256	11	natural	natural	ADJ
cana-5407	256	12	-	-	PUNCT
cana-5407	256	13	sounding	sound	VERB
cana-5407	256	14	speech	speech	NOUN
cana-5407	256	15	while	while	SCONJ
cana-5407	256	16	maintaining	maintain	VERB
cana-5407	256	17	the	the	DET
cana-5407	256	18	unique	unique	ADJ
cana-5407	256	19	characteristics	characteristic	NOUN
cana-5407	256	20	of	of	ADP
cana-5407	256	21	the	the	DET
cana-5407	256	22	target	target	NOUN
cana-5407	256	23	voice	voice	NOUN
cana-5407	256	24	.	.	PUNCT
cana-5407	257	1	significant	significant	ADJ
cana-5407	257	2	advancements	advancement	NOUN
cana-5407	257	3	were	be	AUX
cana-5407	257	4	observed	observe	VERB
cana-5407	257	5	in	in	ADP
cana-5407	257	6	the	the	DET
cana-5407	257	7	quality	quality	NOUN
cana-5407	257	8	of	of	ADP
cana-5407	257	9	voice	voice	NOUN
cana-5407	257	10	conversion	conversion	NOUN
cana-5407	257	11	,	,	PUNCT
cana-5407	257	12	attributed	attribute	VERB
cana-5407	257	13	to	to	ADP
cana-5407	257	14	the	the	DET
cana-5407	257	15	robust	robust	ADJ
cana-5407	257	16	feature	feature	NOUN
cana-5407	257	17	extraction	extraction	NOUN
cana-5407	257	18	capabilities	capability	NOUN
cana-5407	257	19	of	of	ADP
cana-5407	257	20	the	the	DET
cana-5407	257	21	encoder	encoder	NOUN
cana-5407	257	22	and	and	CCONJ
cana-5407	257	23	the	the	DET
cana-5407	257	24	precise	precise	ADJ
cana-5407	257	25	generation	generation	NOUN
cana-5407	257	26	qualities	quality	NOUN
cana-5407	257	27	of	of	ADP
cana-5407	257	28	the	the	DET
cana-5407	257	29	decoder	decoder	NOUN
cana-5407	257	30	.	.	PUNCT
cana-5407	258	1	the	the	DET
cana-5407	258	2	application	application	NOUN
cana-5407	258	3	of	of	ADP
cana-5407	258	4	bidirectional	bidirectional	ADJ
cana-5407	258	5	lstms	lstms	NOUN
cana-5407	258	6	and	and	CCONJ
cana-5407	258	7	wavenet	wavenet	ADJ
cana-5407	258	8	vocoder	vocoder	NOUN
cana-5407	258	9	further	far	ADV
cana-5407	258	10	refined	refine	VERB
cana-5407	258	11	the	the	DET
cana-5407	258	12	audio	audio	ADJ
cana-5407	258	13	output	output	NOUN
cana-5407	258	14	,	,	PUNCT
cana-5407	258	15	ensuring	ensure	VERB
cana-5407	258	16	high	high	ADJ
cana-5407	258	17	fidelity	fidelity	NOUN
cana-5407	258	18	and	and	CCONJ
cana-5407	258	19	clarity	clarity	NOUN
cana-5407	258	20	.	.	PUNCT
cana-5407	259	1	overall	overall	ADV
cana-5407	259	2	,	,	PUNCT
cana-5407	259	3	the	the	DET
cana-5407	259	4	proposed	propose	VERB
cana-5407	259	5	model	model	NOUN
cana-5407	259	6	not	not	PART
cana-5407	259	7	only	only	ADV
cana-5407	259	8	demonstrates	demonstrate	VERB
cana-5407	259	9	superior	superior	ADJ
cana-5407	259	10	performance	performance	NOUN
cana-5407	259	11	in	in	ADP
cana-5407	259	12	voice	voice	NOUN
cana-5407	259	13	conversion	conversion	NOUN
cana-5407	259	14	but	but	CCONJ
cana-5407	259	15	also	also	ADV
cana-5407	259	16	offers	offer	VERB
cana-5407	259	17	new	new	ADJ
cana-5407	259	18	directions	direction	NOUN
cana-5407	259	19	for	for	ADP
cana-5407	259	20	future	future	ADJ
cana-5407	259	21	research	research	NOUN
cana-5407	259	22	in	in	ADP
cana-5407	259	23	enhancing	enhance	VERB
cana-5407	259	24	speech	speech	NOUN
cana-5407	259	25	synthesis	synthesis	NOUN
cana-5407	259	26	technologies	technology	NOUN
cana-5407	259	27	.	.	PUNCT
cana-5407	260	1	references	reference	NOUN
cana-5407	260	2	:	:	PUNCT
cana-5407	261	1	[	[	X
cana-5407	261	2	1	1	NUM
cana-5407	261	3	]	]	SYM
cana-5407	261	4	yoo	yoo	NOUN
cana-5407	261	5	,	,	PUNCT
cana-5407	261	6	in	in	ADP
cana-5407	261	7	-	-	PUNCT
cana-5407	261	8	chul	chul	NOUN
cana-5407	261	9	,	,	PUNCT
cana-5407	261	10	keonnyeong	keonnyeong	PROPN
cana-5407	261	11	lee	lee	PROPN
cana-5407	261	12	,	,	PUNCT
cana-5407	261	13	seonggyun	seonggyun	PROPN
cana-5407	261	14	leem	leem	PROPN
cana-5407	261	15	,	,	PUNCT
cana-5407	261	16	hyunwoo	hyunwoo	NOUN
cana-5407	261	17	oh	oh	INTJ
cana-5407	261	18	,	,	PUNCT
cana-5407	261	19	bonggu	bonggu	PROPN
cana-5407	261	20	ko	ko	PROPN
cana-5407	261	21	,	,	PUNCT
cana-5407	261	22	and	and	CCONJ
cana-5407	261	23	dongsuk	dongsuk	NOUN
cana-5407	261	24	yook	yook	NOUN
cana-5407	261	25	.	.	PUNCT
cana-5407	262	1	"	"	PUNCT
cana-5407	262	2	speaker	speaker	NOUN
cana-5407	262	3	anonymization	anonymization	NOUN
cana-5407	262	4	for	for	ADP
cana-5407	262	5	personal	personal	ADJ
cana-5407	262	6	information	information	NOUN
cana-5407	262	7	protection	protection	NOUN
cana-5407	262	8	using	use	VERB
cana-5407	262	9	voice	voice	NOUN
cana-5407	262	10	conversion	conversion	NOUN
cana-5407	262	11	techniques	technique	NOUN
cana-5407	262	12	.	.	PUNCT
cana-5407	262	13	"	"	PUNCT
cana-5407	263	1	ieee	ieee	NOUN
cana-5407	263	2	access	access	NOUN
cana-5407	263	3	8	8	NUM
cana-5407	263	4	(	(	PUNCT
cana-5407	263	5	2020	2020	NUM
cana-5407	263	6	):	):	PUNCT
cana-5407	263	7	198637	198637	NUM
cana-5407	263	8	-	-	SYM
cana-5407	263	9	198645	198645	NUM
cana-5407	263	10	.	.	PUNCT
cana-5407	264	1	[	[	X
cana-5407	264	2	2	2	NUM
cana-5407	264	3	]	]	X
cana-5407	264	4	polyak	polyak	NOUN
cana-5407	264	5	,	,	PUNCT
cana-5407	264	6	adam	adam	PROPN
cana-5407	264	7	,	,	PUNCT
cana-5407	264	8	lior	lior	PROPN
cana-5407	264	9	wolf	wolf	PROPN
cana-5407	264	10	,	,	PUNCT
cana-5407	264	11	yossi	yossi	PROPN
cana-5407	264	12	adi	adi	PROPN
cana-5407	264	13	,	,	PUNCT
cana-5407	264	14	and	and	CCONJ
cana-5407	264	15	yaniv	yaniv	PROPN
cana-5407	264	16	taigman	taigman	NOUN
cana-5407	264	17	.	.	PUNCT
cana-5407	265	1	"	"	PUNCT
cana-5407	265	2	unsupervised	unsupervised	ADJ
cana-5407	265	3	cross	cross	ADJ
cana-5407	265	4	-	-	ADJ
cana-5407	265	5	domain	domain	ADJ
cana-5407	265	6	singing	singing	NOUN
cana-5407	265	7	voice	voice	NOUN
cana-5407	265	8	conversion	conversion	NOUN
cana-5407	265	9	.	.	PUNCT
cana-5407	265	10	"	"	PUNCT
cana-5407	266	1	arxiv	arxiv	PROPN
cana-5407	266	2	preprint	preprint	NOUN
cana-5407	266	3	arxiv:2008.02830	arxiv:2008.02830	NOUN
cana-5407	266	4	(	(	PUNCT
cana-5407	266	5	2020	2020	NUM
cana-5407	266	6	)	)	PUNCT
cana-5407	266	7	.	.	PUNCT
cana-5407	267	1	[	[	X
cana-5407	267	2	3	3	X
cana-5407	267	3	]	]	X
cana-5407	267	4	liu	liu	PROPN
cana-5407	267	5	,	,	PUNCT
cana-5407	267	6	songxiang	songxiang	PROPN
cana-5407	267	7	,	,	PUNCT
cana-5407	267	8	yuewen	yuewen	NOUN
cana-5407	267	9	cao	cao	PROPN
cana-5407	267	10	,	,	PUNCT
cana-5407	267	11	shiyin	shiyin	PROPN
cana-5407	267	12	kang	kang	PROPN
cana-5407	267	13	,	,	PUNCT
cana-5407	267	14	na	na	ADP
cana-5407	267	15	hu	hu	PROPN
cana-5407	267	16	,	,	PUNCT
cana-5407	267	17	xunying	xunye	VERB
cana-5407	267	18	liu	liu	PROPN
cana-5407	267	19	,	,	PUNCT
cana-5407	267	20	dan	dan	PROPN
cana-5407	267	21	su	su	PROPN
cana-5407	267	22	,	,	PUNCT
cana-5407	267	23	dong	dong	PROPN
cana-5407	267	24	yu	yu	PROPN
cana-5407	267	25	,	,	PUNCT
cana-5407	267	26	and	and	CCONJ
cana-5407	267	27	helen	helen	PROPN
cana-5407	267	28	meng	meng	PROPN
cana-5407	267	29	.	.	PUNCT
cana-5407	268	1	"	"	PUNCT
cana-5407	268	2	transferring	transfer	VERB
cana-5407	268	3	source	source	NOUN
cana-5407	268	4	style	style	NOUN
cana-5407	268	5	in	in	ADP
cana-5407	268	6	non	non	ADJ
cana-5407	268	7	-	-	ADJ
cana-5407	268	8	parallel	parallel	ADJ
cana-5407	268	9	voice	voice	NOUN
cana-5407	268	10	conversion	conversion	NOUN
cana-5407	268	11	.	.	PUNCT
cana-5407	268	12	"	"	PUNCT
cana-5407	269	1	arxiv	arxiv	PROPN
cana-5407	269	2	preprint	preprint	VERB
cana-5407	269	3	arxiv:2005.09178	arxiv:2005.09178	ADV
cana-5407	269	4	(	(	PUNCT
cana-5407	269	5	2020	2020	NUM
cana-5407	269	6	)	)	PUNCT
cana-5407	269	7	.	.	PUNCT
cana-5407	270	1	[	[	X
cana-5407	270	2	4	4	NUM
cana-5407	270	3	]	]	SYM
cana-5407	270	4	deng	deng	PROPN
cana-5407	270	5	,	,	PUNCT
cana-5407	270	6	chengqi	chengqi	PROPN
cana-5407	270	7	,	,	PUNCT
cana-5407	270	8	chengzhu	chengzhu	PROPN
cana-5407	270	9	yu	yu	PROPN
cana-5407	270	10	,	,	PUNCT
cana-5407	270	11	heng	heng	PROPN
cana-5407	270	12	lu	lu	PROPN
cana-5407	270	13	,	,	PUNCT
cana-5407	270	14	chao	chao	PROPN
cana-5407	270	15	weng	weng	PROPN
cana-5407	270	16	,	,	PUNCT
cana-5407	270	17	and	and	CCONJ
cana-5407	270	18	dong	dong	PROPN
cana-5407	270	19	yu	yu	PROPN
cana-5407	270	20	.	.	PUNCT
cana-5407	271	1	"	"	PUNCT
cana-5407	271	2	pitchnet	pitchnet	NOUN
cana-5407	271	3	:	:	PUNCT
cana-5407	271	4	unsupervised	unsupervised	ADJ
cana-5407	271	5	singing	singing	NOUN
cana-5407	271	6	voice	voice	NOUN
cana-5407	271	7	conversion	conversion	NOUN
cana-5407	271	8	with	with	ADP
cana-5407	271	9	pitch	pitch	NOUN
cana-5407	271	10	adversarial	adversarial	ADJ
cana-5407	271	11	network	network	NOUN
cana-5407	271	12	.	.	PUNCT
cana-5407	271	13	"	"	PUNCT
cana-5407	272	1	in	in	ADP
cana-5407	272	2	icassp	icassp	PROPN
cana-5407	272	3	2020	2020	NUM
cana-5407	272	4	-	-	SYM
cana-5407	272	5	2020	2020	NUM
cana-5407	272	6	ieee	ieee	NOUN
cana-5407	272	7	international	international	ADJ
cana-5407	272	8	conference	conference	NOUN
cana-5407	272	9	on	on	ADP
cana-5407	272	10	acoustics	acoustic	NOUN
cana-5407	272	11	,	,	PUNCT
cana-5407	272	12	speech	speech	NOUN
cana-5407	272	13	and	and	CCONJ
cana-5407	272	14	signal	signal	NOUN
cana-5407	272	15	processing	processing	NOUN
cana-5407	272	16	(	(	PUNCT
cana-5407	272	17	icassp	icassp	PROPN
cana-5407	272	18	)	)	PUNCT
cana-5407	272	19	,	,	PUNCT
cana-5407	272	20	pp	pp	ADP
cana-5407	272	21	.	.	PUNCT
cana-5407	273	1	77497753	77497753	NUM
cana-5407	273	2	.	.	PUNCT
cana-5407	274	1	ieee	ieee	NOUN
cana-5407	274	2	,	,	PUNCT
cana-5407	274	3	2020	2020	NUM
cana-5407	274	4	.	.	PUNCT
cana-5407	275	1	[	[	X
cana-5407	275	2	5	5	NUM
cana-5407	275	3	]	]	PUNCT
cana-5407	275	4	choi	choi	NOUN
cana-5407	275	5	,	,	PUNCT
cana-5407	275	6	heejin	heejin	ADJ
cana-5407	275	7	,	,	PUNCT
cana-5407	275	8	and	and	CCONJ
cana-5407	275	9	minsoo	minsoo	ADJ
cana-5407	275	10	hahn	hahn	PROPN
cana-5407	275	11	.	.	PUNCT
cana-5407	276	1	"	"	PUNCT
cana-5407	276	2	sequence	sequence	NOUN
cana-5407	276	3	-	-	PUNCT
cana-5407	276	4	to	to	ADP
cana-5407	276	5	-	-	PUNCT
cana-5407	276	6	sequence	sequence	NOUN
cana-5407	276	7	emotional	emotional	ADJ
cana-5407	276	8	voice	voice	NOUN
cana-5407	276	9	conversion	conversion	NOUN
cana-5407	276	10	with	with	ADP
cana-5407	276	11	strength	strength	NOUN
cana-5407	276	12	control	control	NOUN
cana-5407	276	13	.	.	PUNCT
cana-5407	276	14	"	"	PUNCT
cana-5407	277	1	ieee	ieee	NOUN
cana-5407	277	2	access	access	NOUN
cana-5407	277	3	9	9	NUM
cana-5407	277	4	(	(	PUNCT
cana-5407	277	5	2021	2021	NUM
cana-5407	277	6	):	):	PUNCT
cana-5407	277	7	42674	42674	NUM
cana-5407	277	8	-	-	SYM
cana-5407	277	9	42687	42687	NUM
cana-5407	277	10	.	.	PUNCT
cana-5407	278	1	[	[	X
cana-5407	278	2	6	6	NUM
cana-5407	278	3	]	]	X
cana-5407	278	4	zhang	zhang	PROPN
cana-5407	278	5	,	,	PUNCT
cana-5407	278	6	liqiang	liqiang	PROPN
cana-5407	278	7	,	,	PUNCT
cana-5407	278	8	chengzhu	chengzhu	PROPN
cana-5407	278	9	yu	yu	PROPN
cana-5407	278	10	,	,	PUNCT
cana-5407	278	11	heng	heng	PROPN
cana-5407	278	12	lu	lu	PROPN
cana-5407	278	13	,	,	PUNCT
cana-5407	278	14	chao	chao	PROPN
cana-5407	278	15	weng	weng	PROPN
cana-5407	278	16	,	,	PUNCT
cana-5407	278	17	chunlei	chunlei	VERB
cana-5407	278	18	zhang	zhang	PROPN
cana-5407	278	19	,	,	PUNCT
cana-5407	278	20	yusong	yusong	PROPN
cana-5407	278	21	wu	wu	PROPN
cana-5407	278	22	,	,	PUNCT
cana-5407	278	23	xiang	xiang	PROPN
cana-5407	278	24	xie	xie	PROPN
cana-5407	278	25	,	,	PUNCT
cana-5407	278	26	zijin	zijin	PROPN
cana-5407	278	27	li	li	PROPN
cana-5407	278	28	,	,	PUNCT
cana-5407	278	29	and	and	CCONJ
cana-5407	278	30	dong	dong	PROPN
cana-5407	278	31	yu	yu	PROPN
cana-5407	278	32	.	.	PUNCT
cana-5407	279	1	"	"	PUNCT
cana-5407	279	2	durian	durian	PROPN
cana-5407	279	3	-	-	PUNCT
cana-5407	279	4	sc	sc	NOUN
cana-5407	279	5	:	:	PUNCT
cana-5407	279	6	duration	duration	NOUN
cana-5407	279	7	informed	inform	VERB
cana-5407	279	8	attention	attention	NOUN
cana-5407	279	9	network	network	NOUN
cana-5407	279	10	based	base	VERB
cana-5407	279	11	singing	singing	NOUN
cana-5407	279	12	voice	voice	NOUN
cana-5407	279	13	conversion	conversion	NOUN
cana-5407	279	14	system	system	NOUN
cana-5407	279	15	.	.	PUNCT
cana-5407	279	16	"	"	PUNCT
cana-5407	280	1	arxiv	arxiv	PROPN
cana-5407	280	2	preprint	preprint	NOUN
cana-5407	280	3	arxiv:2008.03009	arxiv:2008.03009	PROPN
cana-5407	280	4	(	(	PUNCT
cana-5407	280	5	2020	2020	NUM
cana-5407	280	6	)	)	PUNCT
cana-5407	280	7	.	.	PUNCT
cana-5407	281	1	communications	communication	NOUN
cana-5407	281	2	on	on	ADP
cana-5407	281	3	applied	apply	VERB
cana-5407	281	4	nonlinear	nonlinear	ADJ
cana-5407	281	5	analysis	analysis	NOUN
cana-5407	281	6	issn	issn	NOUN
cana-5407	281	7	:	:	PUNCT
cana-5407	281	8	1074	1074	NUM
cana-5407	281	9	-	-	PUNCT
cana-5407	281	10	133x	133x	NUM
cana-5407	281	11	vol	vol	VERB
cana-5407	281	12	32	32	NUM
cana-5407	281	13	no	no	NOUN
cana-5407	281	14	.	.	PUNCT
cana-5407	282	1	10s	10	NOUN
cana-5407	282	2	(	(	PUNCT
cana-5407	282	3	2025	2025	NUM
cana-5407	282	4	)	)	PUNCT
cana-5407	282	5	2185	2185	NUM
cana-5407	282	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-5407	283	1	[	[	X
cana-5407	283	2	7	7	NUM
cana-5407	283	3	]	]	PUNCT
cana-5407	283	4	baas	baas	NOUN
cana-5407	283	5	,	,	PUNCT
cana-5407	283	6	matthew	matthew	PROPN
cana-5407	283	7	,	,	PUNCT
cana-5407	283	8	and	and	CCONJ
cana-5407	283	9	herman	herman	PROPN
cana-5407	283	10	kamper	kamper	PROPN
cana-5407	283	11	.	.	PUNCT
cana-5407	284	1	"	"	PUNCT
cana-5407	284	2	voice	voice	NOUN
cana-5407	284	3	conversion	conversion	NOUN
cana-5407	284	4	for	for	ADP
cana-5407	284	5	stuttered	stuttered	ADJ
cana-5407	284	6	speech	speech	NOUN
cana-5407	284	7	,	,	PUNCT
cana-5407	284	8	instruments	instrument	NOUN
cana-5407	284	9	,	,	PUNCT
cana-5407	284	10	unseen	unseen	ADJ
cana-5407	284	11	languages	language	NOUN
cana-5407	284	12	and	and	CCONJ
cana-5407	284	13	textually	textually	ADV
cana-5407	284	14	described	describe	VERB
cana-5407	284	15	voices	voice	NOUN
cana-5407	284	16	.	.	PUNCT
cana-5407	284	17	"	"	PUNCT
cana-5407	285	1	in	in	ADP
cana-5407	285	2	southern	southern	ADJ
cana-5407	285	3	african	african	ADJ
cana-5407	285	4	conference	conference	NOUN
cana-5407	285	5	for	for	ADP
cana-5407	285	6	artificial	artificial	ADJ
cana-5407	285	7	intelligence	intelligence	NOUN
cana-5407	285	8	research	research	NOUN
cana-5407	285	9	,	,	PUNCT
cana-5407	285	10	pp	pp	ADJ
cana-5407	285	11	.	.	PUNCT
cana-5407	286	1	136	136	NUM
cana-5407	286	2	-	-	SYM
cana-5407	286	3	150	150	NUM
cana-5407	286	4	.	.	PUNCT
cana-5407	287	1	cham	cham	PROPN
cana-5407	287	2	:	:	PUNCT
cana-5407	287	3	springer	springer	NOUN
cana-5407	287	4	nature	nature	PROPN
cana-5407	287	5	switzerland	switzerland	PROPN
cana-5407	287	6	,	,	PUNCT
cana-5407	287	7	2023	2023	NUM
cana-5407	287	8	.	.	PUNCT
cana-5407	288	1	[	[	X
cana-5407	288	2	8	8	NUM
cana-5407	288	3	]	]	X
cana-5407	288	4	zheng	zheng	X
cana-5407	288	5	,	,	PUNCT
cana-5407	288	6	wei	wei	PROPN
cana-5407	288	7	-	-	PROPN
cana-5407	288	8	zhong	zhong	PROPN
cana-5407	288	9	,	,	PUNCT
cana-5407	288	10	ji	ji	PROPN
cana-5407	288	11	-	-	PUNCT
cana-5407	288	12	yan	yan	PROPN
cana-5407	288	13	han	han	PROPN
cana-5407	288	14	,	,	PUNCT
cana-5407	288	15	chen	chen	PROPN
cana-5407	288	16	-	-	PUNCT
cana-5407	288	17	kai	kai	PROPN
cana-5407	288	18	lee	lee	PROPN
cana-5407	288	19	,	,	PUNCT
cana-5407	288	20	yu	yu	PROPN
cana-5407	288	21	-	-	PUNCT
cana-5407	288	22	yi	yi	PROPN
cana-5407	288	23	lin	lin	PROPN
cana-5407	288	24	,	,	PUNCT
cana-5407	288	25	shu	shu	PROPN
cana-5407	288	26	-	-	PUNCT
cana-5407	288	27	han	han	PROPN
cana-5407	288	28	chang	chang	PROPN
cana-5407	288	29	,	,	PUNCT
cana-5407	288	30	and	and	CCONJ
cana-5407	288	31	ying	ying	PROPN
cana-5407	288	32	-	-	PUNCT
cana-5407	288	33	hui	hui	PROPN
cana-5407	288	34	lai	lai	PROPN
cana-5407	288	35	.	.	PUNCT
cana-5407	289	1	"	"	PUNCT
cana-5407	289	2	phonetic	phonetic	ADJ
cana-5407	289	3	posteriorgram	posteriorgram	NOUN
cana-5407	289	4	-	-	PUNCT
cana-5407	289	5	based	base	VERB
cana-5407	289	6	voice	voice	NOUN
cana-5407	289	7	conversion	conversion	NOUN
cana-5407	289	8	system	system	NOUN
cana-5407	289	9	to	to	PART
cana-5407	289	10	improve	improve	VERB
cana-5407	289	11	speech	speech	NOUN
cana-5407	289	12	intelligibility	intelligibility	NOUN
cana-5407	289	13	of	of	ADP
cana-5407	289	14	dysarthric	dysarthric	ADJ
cana-5407	289	15	patients	patient	NOUN
cana-5407	289	16	.	.	PUNCT
cana-5407	289	17	"	"	PUNCT
cana-5407	290	1	computer	computer	NOUN
cana-5407	290	2	methods	method	NOUN
cana-5407	290	3	and	and	CCONJ
cana-5407	290	4	programs	program	NOUN
cana-5407	290	5	in	in	ADP
cana-5407	290	6	biomedicine	biomedicine	NOUN
cana-5407	290	7	215	215	NUM
cana-5407	290	8	(	(	PUNCT
cana-5407	290	9	2022	2022	NUM
cana-5407	290	10	):	):	PUNCT
cana-5407	290	11	106602	106602	NUM
cana-5407	290	12	.	.	PUNCT
cana-5407	291	1	[	[	X
cana-5407	291	2	9	9	NUM
cana-5407	291	3	]	]	SYM
cana-5407	291	4	sisman	sisman	NOUN
cana-5407	291	5	,	,	PUNCT
cana-5407	291	6	berrak	berrak	NOUN
cana-5407	291	7	,	,	PUNCT
cana-5407	291	8	junichi	junichi	PROPN
cana-5407	291	9	yamagishi	yamagishi	PROPN
cana-5407	291	10	,	,	PUNCT
cana-5407	291	11	simon	simon	PROPN
cana-5407	291	12	king	king	PROPN
cana-5407	291	13	,	,	PUNCT
cana-5407	291	14	and	and	CCONJ
cana-5407	291	15	haizhou	haizhou	PROPN
cana-5407	291	16	li	li	PROPN
cana-5407	291	17	.	.	PUNCT
cana-5407	292	1	"	"	PUNCT
cana-5407	292	2	an	an	DET
cana-5407	292	3	overview	overview	NOUN
cana-5407	292	4	of	of	ADP
cana-5407	292	5	voice	voice	NOUN
cana-5407	292	6	conversion	conversion	NOUN
cana-5407	292	7	and	and	CCONJ
cana-5407	292	8	its	its	PRON
cana-5407	292	9	challenges	challenge	NOUN
cana-5407	292	10	:	:	PUNCT
cana-5407	292	11	from	from	ADP
cana-5407	292	12	statistical	statistical	ADJ
cana-5407	292	13	modeling	modeling	NOUN
cana-5407	292	14	to	to	ADP
cana-5407	292	15	deep	deep	ADJ
cana-5407	292	16	learning	learning	NOUN
cana-5407	292	17	.	.	PUNCT
cana-5407	292	18	"	"	PUNCT
cana-5407	292	19	ieee	ieee	PROPN
cana-5407	292	20	/	/	SYM
cana-5407	292	21	acm	acm	NOUN
cana-5407	292	22	transactions	transaction	NOUN
cana-5407	292	23	on	on	ADP
cana-5407	292	24	audio	audio	NOUN
cana-5407	292	25	,	,	PUNCT
cana-5407	292	26	speech	speech	NOUN
cana-5407	292	27	,	,	PUNCT
cana-5407	292	28	and	and	CCONJ
cana-5407	292	29	language	language	NOUN
cana-5407	292	30	processing	processing	NOUN
cana-5407	292	31	29	29	NUM
cana-5407	292	32	(	(	PUNCT
cana-5407	292	33	2020	2020	NUM
cana-5407	292	34	):	):	PUNCT
cana-5407	292	35	132	132	NUM
cana-5407	292	36	-	-	SYM
cana-5407	292	37	157	157	NUM
cana-5407	292	38	.	.	PUNCT
cana-5407	293	1	[	[	X
cana-5407	293	2	10	10	NUM
cana-5407	293	3	]	]	X
cana-5407	293	4	zhao	zhao	X
cana-5407	293	5	,	,	PUNCT
cana-5407	293	6	yi	yi	PROPN
cana-5407	293	7	,	,	PUNCT
cana-5407	293	8	wen	wen	PROPN
cana-5407	293	9	-	-	PROPN
cana-5407	293	10	chin	chin	PROPN
cana-5407	293	11	huang	huang	PROPN
cana-5407	293	12	,	,	PUNCT
cana-5407	293	13	xiaohai	xiaohai	PROPN
cana-5407	293	14	tian	tian	PROPN
cana-5407	293	15	,	,	PUNCT
cana-5407	293	16	junichi	junichi	PROPN
cana-5407	293	17	yamagishi	yamagishi	PROPN
cana-5407	293	18	,	,	PUNCT
cana-5407	293	19	rohan	rohan	PROPN
cana-5407	293	20	kumar	kumar	PROPN
cana-5407	293	21	das	das	PROPN
cana-5407	293	22	,	,	PUNCT
cana-5407	293	23	tomi	tomi	PROPN
cana-5407	293	24	kinnunen	kinnunen	PROPN
cana-5407	293	25	,	,	PUNCT
cana-5407	293	26	zhenhua	zhenhua	PROPN
cana-5407	293	27	ling	ling	PROPN
cana-5407	293	28	,	,	PUNCT
cana-5407	293	29	and	and	CCONJ
cana-5407	293	30	tomoki	tomoki	NOUN
cana-5407	293	31	toda	toda	NOUN
cana-5407	293	32	.	.	PUNCT
cana-5407	294	1	"	"	PUNCT
cana-5407	294	2	voice	voice	NOUN
cana-5407	294	3	conversion	conversion	NOUN
cana-5407	294	4	challenge	challenge	NOUN
cana-5407	294	5	2020	2020	NUM
cana-5407	294	6	:	:	PUNCT
cana-5407	294	7	intra	intra	ADJ
cana-5407	294	8	-	-	ADJ
cana-5407	294	9	lingual	lingual	ADJ
cana-5407	294	10	semi	semi	ADJ
cana-5407	294	11	-	-	ADJ
cana-5407	294	12	parallel	parallel	ADJ
cana-5407	294	13	and	and	CCONJ
cana-5407	294	14	cross	cross	ADJ
cana-5407	294	15	-	-	ADJ
cana-5407	294	16	lingual	lingual	ADJ
cana-5407	294	17	voice	voice	NOUN
cana-5407	294	18	conversion	conversion	NOUN
cana-5407	294	19	.	.	PUNCT
cana-5407	294	20	"	"	PUNCT
cana-5407	295	1	arxiv	arxiv	PROPN
cana-5407	295	2	preprint	preprint	NOUN
cana-5407	295	3	arxiv:2008.12527	arxiv:2008.12527	NOUN
cana-5407	295	4	(	(	PUNCT
cana-5407	295	5	2020	2020	NUM
cana-5407	295	6	)	)	PUNCT
cana-5407	295	7	.	.	PUNCT
cana-5407	296	1	[	[	X
cana-5407	296	2	11	11	NUM
cana-5407	296	3	]	]	X
cana-5407	296	4	zhou	zhou	PROPN
cana-5407	296	5	,	,	PUNCT
cana-5407	296	6	kun	kun	PROPN
cana-5407	296	7	,	,	PUNCT
cana-5407	296	8	berrak	berrak	PROPN
cana-5407	296	9	sisman	sisman	NOUN
cana-5407	296	10	,	,	PUNCT
cana-5407	296	11	rui	rui	PROPN
cana-5407	296	12	liu	liu	PROPN
cana-5407	296	13	,	,	PUNCT
cana-5407	296	14	and	and	CCONJ
cana-5407	296	15	haizhou	haizhou	PROPN
cana-5407	296	16	li	li	PROPN
cana-5407	296	17	.	.	PUNCT
cana-5407	297	1	"	"	PUNCT
cana-5407	297	2	emotional	emotional	ADJ
cana-5407	297	3	voice	voice	NOUN
cana-5407	297	4	conversion	conversion	NOUN
cana-5407	297	5	:	:	PUNCT
cana-5407	298	1	theory	theory	NOUN
cana-5407	298	2	,	,	PUNCT
cana-5407	298	3	databases	database	NOUN
cana-5407	298	4	and	and	CCONJ
cana-5407	298	5	esd	esd	NOUN
cana-5407	298	6	.	.	PUNCT
cana-5407	298	7	"	"	PUNCT
cana-5407	298	8	speech	speech	NOUN
cana-5407	298	9	communication	communication	NOUN
cana-5407	298	10	137	137	NUM
cana-5407	298	11	(	(	PUNCT
cana-5407	298	12	2022	2022	NUM
cana-5407	298	13	):	):	PUNCT
cana-5407	298	14	1	1	NUM
cana-5407	298	15	-	-	SYM
cana-5407	298	16	18	18	NUM
cana-5407	298	17	.	.	PUNCT
cana-5407	299	1	[	[	X
cana-5407	299	2	12	12	NUM
cana-5407	299	3	]	]	X
cana-5407	299	4	wang	wang	PROPN
cana-5407	299	5	,	,	PUNCT
cana-5407	299	6	disong	disong	PROPN
cana-5407	299	7	,	,	PUNCT
cana-5407	299	8	liqun	liqun	PROPN
cana-5407	299	9	deng	deng	PROPN
cana-5407	299	10	,	,	PUNCT
cana-5407	299	11	yu	yu	PROPN
cana-5407	299	12	ting	ting	PROPN
cana-5407	299	13	yeung	yeung	PROPN
cana-5407	299	14	,	,	PUNCT
cana-5407	299	15	xiao	xiao	PROPN
cana-5407	299	16	chen	chen	PROPN
cana-5407	299	17	,	,	PUNCT
cana-5407	299	18	xunying	xunye	VERB
cana-5407	299	19	liu	liu	PROPN
cana-5407	299	20	,	,	PUNCT
cana-5407	299	21	and	and	CCONJ
cana-5407	299	22	helen	helen	PROPN
cana-5407	299	23	meng	meng	PROPN
cana-5407	299	24	.	.	PUNCT
cana-5407	300	1	"	"	PUNCT
cana-5407	300	2	vqmivc	vqmivc	NOUN
cana-5407	300	3	:	:	PUNCT
cana-5407	300	4	vector	vector	NOUN
cana-5407	300	5	quantization	quantization	NOUN
cana-5407	300	6	and	and	CCONJ
cana-5407	300	7	mutual	mutual	ADJ
cana-5407	300	8	information	information	NOUN
cana-5407	300	9	-	-	PUNCT
cana-5407	300	10	based	base	VERB
cana-5407	300	11	unsupervised	unsupervised	ADJ
cana-5407	300	12	speech	speech	NOUN
cana-5407	300	13	representation	representation	NOUN
cana-5407	300	14	disentanglement	disentanglement	NOUN
cana-5407	300	15	for	for	ADP
cana-5407	300	16	one	one	NUM
cana-5407	300	17	-	-	PUNCT
cana-5407	300	18	shot	shot	NOUN
cana-5407	300	19	voice	voice	NOUN
cana-5407	300	20	conversion	conversion	NOUN
cana-5407	300	21	.	.	PUNCT
cana-5407	300	22	"	"	PUNCT
cana-5407	301	1	arxiv	arxiv	PROPN
cana-5407	301	2	preprint	preprint	VERB
cana-5407	301	3	arxiv:2106.10132	arxiv:2106.10132	X
cana-5407	301	4	(	(	PUNCT
cana-5407	301	5	2021	2021	NUM
cana-5407	301	6	)	)	PUNCT
cana-5407	301	7	.	.	PUNCT
cana-5407	302	1	[	[	X
cana-5407	302	2	13	13	NUM
cana-5407	302	3	]	]	SYM
cana-5407	302	4	li	li	PROPN
cana-5407	302	5	,	,	PUNCT
cana-5407	302	6	yinghao	yinghao	PROPN
cana-5407	302	7	aaron	aaron	PROPN
cana-5407	302	8	,	,	PUNCT
cana-5407	302	9	ali	ali	PROPN
cana-5407	302	10	zare	zare	PROPN
cana-5407	302	11	,	,	PUNCT
cana-5407	302	12	and	and	CCONJ
cana-5407	302	13	nima	nima	PROPN
cana-5407	302	14	mesgarani	mesgarani	PROPN
cana-5407	302	15	.	.	PUNCT
cana-5407	303	1	"	"	PUNCT
cana-5407	303	2	starganv2	starganv2	PROPN
cana-5407	303	3	-	-	PUNCT
cana-5407	303	4	vc	vc	PROPN
cana-5407	303	5	:	:	PUNCT
cana-5407	303	6	a	a	DET
cana-5407	303	7	diverse	diverse	ADJ
cana-5407	303	8	,	,	PUNCT
cana-5407	303	9	unsupervised	unsupervised	ADJ
cana-5407	303	10	,	,	PUNCT
cana-5407	303	11	non	non	ADJ
cana-5407	303	12	-	-	ADJ
cana-5407	303	13	parallel	parallel	ADJ
cana-5407	303	14	framework	framework	NOUN
cana-5407	303	15	for	for	ADP
cana-5407	303	16	natural	natural	ADJ
cana-5407	303	17	-	-	PUNCT
cana-5407	303	18	sounding	sound	VERB
cana-5407	303	19	voice	voice	NOUN
cana-5407	303	20	conversion	conversion	NOUN
cana-5407	303	21	.	.	PUNCT
cana-5407	303	22	"	"	PUNCT
cana-5407	304	1	arxiv	arxiv	PROPN
cana-5407	304	2	preprint	preprint	NOUN
cana-5407	304	3	arxiv:2107.10394	arxiv:2107.10394	NOUN
cana-5407	304	4	(	(	PUNCT
cana-5407	304	5	2021	2021	NUM
cana-5407	304	6	)	)	PUNCT
cana-5407	304	7	.	.	PUNCT
cana-5407	305	1	[	[	X
cana-5407	305	2	14	14	NUM
cana-5407	305	3	]	]	X
cana-5407	305	4	liu	liu	PROPN
cana-5407	305	5	,	,	PUNCT
cana-5407	305	6	songxiang	songxiang	PROPN
cana-5407	305	7	,	,	PUNCT
cana-5407	305	8	yuewen	yuewen	NOUN
cana-5407	305	9	cao	cao	NOUN
cana-5407	305	10	,	,	PUNCT
cana-5407	305	11	disong	disong	PROPN
cana-5407	305	12	wang	wang	PROPN
cana-5407	305	13	,	,	PUNCT
cana-5407	305	14	xixin	xixin	PROPN
cana-5407	305	15	wu	wu	PROPN
cana-5407	305	16	,	,	PUNCT
cana-5407	305	17	xunying	xunye	VERB
cana-5407	305	18	liu	liu	PROPN
cana-5407	305	19	,	,	PUNCT
cana-5407	305	20	and	and	CCONJ
cana-5407	305	21	helen	helen	PROPN
cana-5407	305	22	meng	meng	PROPN
cana-5407	305	23	.	.	PUNCT
cana-5407	306	1	"	"	PUNCT
cana-5407	306	2	anyto	anyto	ADJ
cana-5407	306	3	-	-	PUNCT
cana-5407	306	4	many	many	ADJ
cana-5407	306	5	voice	voice	NOUN
cana-5407	306	6	conversion	conversion	NOUN
cana-5407	306	7	with	with	ADP
cana-5407	306	8	location	location	NOUN
cana-5407	306	9	-	-	PUNCT
cana-5407	306	10	relative	relative	ADJ
cana-5407	306	11	sequence	sequence	NOUN
cana-5407	306	12	-	-	PUNCT
cana-5407	306	13	to	to	ADP
cana-5407	306	14	-	-	PUNCT
cana-5407	306	15	sequence	sequence	NOUN
cana-5407	306	16	modeling	modeling	NOUN
cana-5407	306	17	.	.	PUNCT
cana-5407	306	18	"	"	PUNCT
cana-5407	306	19	ieee	ieee	PROPN
cana-5407	306	20	/	/	SYM
cana-5407	306	21	acm	acm	NOUN
cana-5407	306	22	transactions	transaction	NOUN
cana-5407	306	23	on	on	ADP
cana-5407	306	24	audio	audio	NOUN
cana-5407	306	25	,	,	PUNCT
cana-5407	306	26	speech	speech	NOUN
cana-5407	306	27	,	,	PUNCT
cana-5407	306	28	and	and	CCONJ
cana-5407	306	29	language	language	NOUN
cana-5407	306	30	processing	processing	NOUN
cana-5407	306	31	29	29	NUM
cana-5407	306	32	(	(	PUNCT
cana-5407	306	33	2021	2021	NUM
cana-5407	306	34	):	):	PUNCT
cana-5407	306	35	1717	1717	NUM
cana-5407	306	36	-	-	SYM
cana-5407	306	37	1728	1728	NUM
cana-5407	306	38	.	.	PUNCT
cana-5407	307	1	[	[	X
cana-5407	307	2	15	15	NUM
cana-5407	307	3	]	]	X
cana-5407	307	4	huang	huang	PROPN
cana-5407	307	5	,	,	PUNCT
cana-5407	307	6	wen	wen	PROPN
cana-5407	307	7	-	-	PUNCT
cana-5407	307	8	chin	chin	PROPN
cana-5407	307	9	,	,	PUNCT
cana-5407	307	10	tomoki	tomoki	PROPN
cana-5407	307	11	hayashi	hayashi	PROPN
cana-5407	307	12	,	,	PUNCT
cana-5407	307	13	yi	yi	PROPN
cana-5407	307	14	-	-	PUNCT
cana-5407	307	15	chiao	chiao	PROPN
cana-5407	307	16	wu	wu	PROPN
cana-5407	307	17	,	,	PUNCT
cana-5407	307	18	hirokazu	hirokazu	NOUN
cana-5407	307	19	kameoka	kameoka	NOUN
cana-5407	307	20	,	,	PUNCT
cana-5407	307	21	and	and	CCONJ
cana-5407	307	22	tomoki	tomoki	NOUN
cana-5407	307	23	toda	toda	PROPN
cana-5407	307	24	.	.	PUNCT
cana-5407	308	1	"	"	PUNCT
cana-5407	308	2	pretraining	pretraine	VERB
cana-5407	308	3	techniques	technique	NOUN
cana-5407	308	4	for	for	ADP
cana-5407	308	5	sequence	sequence	NOUN
cana-5407	308	6	-	-	PUNCT
cana-5407	308	7	to	to	ADP
cana-5407	308	8	-	-	PUNCT
cana-5407	308	9	sequence	sequence	NOUN
cana-5407	308	10	voice	voice	NOUN
cana-5407	308	11	conversion	conversion	NOUN
cana-5407	308	12	.	.	PUNCT
cana-5407	308	13	"	"	PUNCT
cana-5407	308	14	ieee	ieee	PROPN
cana-5407	308	15	/	/	SYM
cana-5407	308	16	acm	acm	NOUN
cana-5407	308	17	transactions	transaction	NOUN
cana-5407	308	18	on	on	ADP
cana-5407	308	19	audio	audio	NOUN
cana-5407	308	20	,	,	PUNCT
cana-5407	308	21	speech	speech	NOUN
cana-5407	308	22	,	,	PUNCT
cana-5407	308	23	and	and	CCONJ
cana-5407	308	24	language	language	NOUN
cana-5407	308	25	processing	processing	NOUN
cana-5407	308	26	29	29	NUM
cana-5407	308	27	(	(	PUNCT
cana-5407	308	28	2021	2021	NUM
cana-5407	308	29	):	):	PUNCT
cana-5407	308	30	745	745	NUM
cana-5407	308	31	-	-	SYM
cana-5407	308	32	755	755	NUM
cana-5407	308	33	.	.	PUNCT
cana-5407	309	1	[	[	X
cana-5407	309	2	16	16	NUM
cana-5407	309	3	]	]	X
cana-5407	309	4	zhou	zhou	PROPN
cana-5407	309	5	,	,	PUNCT
cana-5407	309	6	kun	kun	PROPN
cana-5407	309	7	,	,	PUNCT
cana-5407	309	8	berrak	berrak	PROPN
cana-5407	309	9	sisman	sisman	NOUN
cana-5407	309	10	,	,	PUNCT
cana-5407	309	11	rajib	rajib	PROPN
cana-5407	309	12	rana	rana	PROPN
cana-5407	309	13	,	,	PUNCT
cana-5407	309	14	björn	björn	PROPN
cana-5407	309	15	w.	w.	PROPN
cana-5407	309	16	schuller	schuller	PROPN
cana-5407	309	17	,	,	PUNCT
cana-5407	309	18	and	and	CCONJ
cana-5407	309	19	haizhou	haizhou	PROPN
cana-5407	309	20	li	li	PROPN
cana-5407	309	21	.	.	PUNCT
cana-5407	310	1	"	"	PUNCT
cana-5407	310	2	emotion	emotion	NOUN
cana-5407	310	3	intensity	intensity	NOUN
cana-5407	310	4	and	and	CCONJ
cana-5407	310	5	its	its	PRON
cana-5407	310	6	control	control	NOUN
cana-5407	310	7	for	for	ADP
cana-5407	310	8	emotional	emotional	ADJ
cana-5407	310	9	voice	voice	NOUN
cana-5407	310	10	conversion	conversion	NOUN
cana-5407	310	11	.	.	PUNCT
cana-5407	310	12	"	"	PUNCT
cana-5407	311	1	ieee	ieee	NOUN
cana-5407	311	2	transactions	transaction	NOUN
cana-5407	311	3	on	on	ADP
cana-5407	311	4	affective	affective	ADJ
cana-5407	311	5	computing	computing	NOUN
cana-5407	311	6	14	14	NUM
cana-5407	311	7	,	,	PUNCT
cana-5407	311	8	no	no	INTJ
cana-5407	311	9	.	.	NOUN
cana-5407	311	10	1	1	NUM
cana-5407	311	11	(	(	PUNCT
cana-5407	311	12	2022	2022	NUM
cana-5407	311	13	):	):	PUNCT
cana-5407	311	14	31	31	NUM
cana-5407	311	15	-	-	SYM
cana-5407	311	16	48	48	NUM
cana-5407	311	17	.	.	PUNCT