






































citation


AsiaCALL Online Journal Received: 15/10/2021 

ISSN 1936-9859; https://asiacall.info/acoj  Revision: 22/11/2021 

Vol. 13, No. 1, 2022  Accepted: 05/12/2021 

pp. 93-121  Online: 06/12/2021 

 

CITATION | Le, P. N., Vu, H. M. L., & Tran, M. N. (2022).  Improving EFL Students’ Intonation in Text Using 

Shadowing Technique with the Implementation of Google Text-to-Speech. AsiaCALL Online Journal, 13(1), 93-

121. EOI: http://eoi.citefactor.org/10.11251/acoj.13.01.006 

 

Improving EFL Students' Intonation In-Text Using Shadowing Technique with 

the Implementation of Google Text-to-Speech 

Nhi Le Phuong1, Linh Vu Hoang Mai1*, Nhat Tran Minh1 

1Hoa Sen University, Vietnam 

Correspondence: Linh Vu Hoang Mai, Hoa Sen University, Vietnam. E-mail: 

linhvuhoang1908@gmail.com 

EOI: http://eoi.citefactor.org/10.11251/acoj.13.01.006 

 

Abstract 

Intonation, a feature of pronunciation that embarks on how an utterance is produced with respect 

to the notion of accent, stress and the rising/falling pitch, has always been considered 

troublesome to second language learners due to its connection to personal emotion. Recently, 

greater attention has been drawn from academia to the investigation of the potential solution. In 

contribution to the existing research, this paper is to shed light on the impact of shadowing 

technique with the aid of spoken text features by Google Text-to-Speech on removing learner's 

flat tone as well as achieving basic English intonation in text, provided that the tool is adopted 

intensively throughout a pre-designed training course. In the study, a pretest-posttest method is 

employed in one same sample, with the pre-scheduled experimental period being the control 

group and post-training being the treatment group. After ten weeks of intensive training program 

following a detailed syllabus, the posttest results, which are computerized using Speech 

Analyzer, indicates that the participants benefit from the training in terms of intonation in the 

text as there is a positive difference in scores achieved by the treatment group, benchmarked 

against the control group. However, due to the rather small size of the population, the study 

failed to generalize its findings. The implementation of spoken texts in Google Text-to-Speech 

tool and Speech Analyzer software will be further discussed in this study. 

Keywords: Intonation in text, default tone, Google Text-to-Speech, text-to-speech, shadowing 

techniques 

 

1. Introduction 

It is indisputable that the English language has become a lingua franca with globally widespread 

use. The demands for language learners, thus, have increased accordingly. Besides mastering 

grammatical structures and patterns as well as conquering a wide range of lexical items, learners 

are now required to learn how to use a language in their own way. Together with other features 

of the English language that form productive competence like speaking, intonation starts to be 

considered personal prominence. Assuredly, intonation is deemed one of the most crucial 

attributes a child perceives at the beginning of his first language (Cruttenden, 1997; Crystal, 

1986; Lieberman, 1986; Mehler et al., 1988; Snow & Balog, 2002). Unlike Vietnamese, 

Mandarin, Thai or Japanese, which deal with prosodic characteristics of lexical tones, the 

https://asiacall.info/acoj
http://eoi.citefactor.org/10.11251/acoj.13.01.006
http://eoi.citefactor.org/10.11251/acoj.13.01.006


https://asiacall.info/acoj Le, P. N., Vu, H. M. L., & Tran, M. N. Vol. 13; No. 1; 2022 

94 
 

English language uses tone partly as a medium of intonation to convey information. In terms of 

English language use, the tonal choices of individual words or sentences do not alter the lexical 

meaning of sentences but their non-lexical meaning (Wells, 2006). As a matter of fact, the 

English intonation system poses a number of complicated issues to its users, yet it plays a 

pivotal role in the language's prosody. For other branches of pronunciation, such as phonetically 

pronouncing an individual sound or word, the drill method, which requires students to say the 

words aloud repeatedly, seems effective. However, it is virtually unreasonable to expect that 

the ability to intone properly can be shaped using a similar approach, for intonation involves 

the speaker's feeling at the moment the utterance is produced. 

Recently, prosodic characteristics of intonation have drawn great attention from academia to 

look for possible solutions to the problem mentioned above. In contribution to the existing 

research, this paper is to shed light on the impact of shadowing technique with the aid of spoken 

text features by Google Translation tool, particularly Google Text-to-Speech, on removing 

learners' inappropriate use of English intonation, provided that the tool is adopted intensively 

throughout a pre-designed training course. In the study, a pretest-posttest method is employed 

in one same sample, with the pre-scheduled experimental period being the control group and 

post-training being the treatment group. After six weeks of intensive training program following 

a detailed syllabus, the posttest results, which are computerized using Speech Analyzer, 

indicates that the participants benefit from the training in terms of intonation in the text as there 

is a positive difference in scores achieved by the treatment group, benchmarked against the 

control group. Besides, the implementation of spoken texts in Google Translation tool and 

Speech Analyzer software will be further discussed in this study. 

 

2. Literature review 

2.1. Defining key terms:  

2.1.1. intonation: 

In this paper, intonation is strikingly marked as the rhythm of speech that effectively supports 

the organization and transmission of messages. It is also one of the most broadly used aspects 

of suprasegmental phenomena which has a huge impact on verbal languages - notably the 

English language system. Intonation carries not only grammatical concepts but also syntactic 

and pragmatic meanings. With that being said, it is reasonable to state that language learners 

who made greater efforts in learning and exploiting intonation in their learning have higher 

proficiency in English (Wennerstorm, 1998). Aside from learning how to master the incredibly 

precise pronunciation of discrete words or sentences, learners' awareness and concerns about 

other prosodic features such as intonation should receive more attention from those in charge 

of L2 teaching and training. As a matter of fact, it would be much better if the students 

themselves were provided with proper intonation training. In Lin, Fan, and Chen's views (1995), 

English learners tend to pay greater attention to pronunciation or lexical items such as 

vocabulary and grammatical structures than intonation and rhythm when listening to English. 

This explains why most of them struggle to catch up with the speed of the listening tapes. 

According to Gilbert (1994), tone and its features enable learners to follow the flow of 

information in spoken English. For Piekering (2004) and Wennerstomi (2004; 1998), a non-

native speaker’s English will be deemed more comprehensible by receivers if the user can 

employ proper intonation structures. 



ACOJ- ISSN 1936-9859 AsiaCALL Online Journal  Vol. 13; No. 1; 2022 

95 

Mainly, English language intonation is generally classified into three systems of hierarchy: 

tonality, tonicity and tone (3Ts) (Wells, 2006). Respectively, tonality can be referred to as the 

position at which speakers choose to cut their utterances into chunks. Appropriate intonation 

patterns should be applied to each chunk. Those chunks are widely known under the names like 

the intonation of word group, tone group or intonation phrases - so-called IPs. Considered the 

next hierarchical matter, tonicity is a term showing that language users highlight the most 

crucial words they desire to express (Wells, 2006). These accented words are expected to 

receive much attention from the hearers. More often than not, the speakers accent the main 

stressed syllable of the word to be more accurate. At the time, they shift the pitch movement by 

rising from low to high tones. This is to say, the last matter in 3Ts happens to be born - tone. 

Particularly, the last and the most important accent in IP is called the nucleus. The tone is 

defined and decided when the speaker locates the nucleus in terms of pitch. The tone of 

sentences is closely associated with the type of pitch movement placed. 

Basically, the most fundamental distinctions of nucleus tones are falling, rising, and falling-

rising tones. In this study, the three tones are again generalized into falls and non-fall since 

falling tones, and their variations are varied and rather vague (Wells, 2006). In falling-toned 

sentences, the tails fall lower than its nucleus and rising-toned sentences with that in vice versa.    

(fall)   I           think              it              was               funny. 

        •                              ● 

∙             ∙     ∙          ∙ 
Nucleus|tail 

(rise)   Was              it              funny? 

∙ 

● 

∙       ∙              
Nucleus|tail 

In the two examples, the syllables -fun- are treated as the nucleuses which are accented strongly 

with appropriate timing, and the syllables -ny- are tails. The tail in (fall) examples goes lower 

than its nucleus compared to that of (rise) example. Therefore, (fall) example is determined as 

falling tone and (rise) example is defined as rising tone. In this paper, the research team 

principally pays huge attention to default tones that are relatively associated with the notion of 

tones regarding each type of sentence. Our target participants are non-English majors who have 

loose English backgrounds and no attachment to the language. As generalized, default tones 

can be: 

● a fall as in sentence types such as statements, commands and wh-questions, 

exclamations; 



https://asiacall.info/acoj Le, P. N., Vu, H. M. L., & Tran, M. N. Vol. 13; No. 1; 2022 

96 
 

● a rise as in questions requiring confirmation such as yes-no questions; 

Apart from that, another generalization of default tones involves the notions of utterances. The 

default can be: 

● a fall if it displays the main part of the utterance, and 

● a fall-rise/rise (non-fall) if it displays a subordinate or dependent clause. 

2.1.2. Google Text-to-Speech (GTTS) 

Google Text-to-Speech (GTTS) with WaveNet occurred as a momentous development of speech 

synthesis technology.    

Google Text-to-Speech (GTTS) is an application of screen reader used for the Android 

operating system, powered by Google. It enables apps to read out loud the texts appearing on 

the screen in a variety of languages. Apps such as Google Play Books for using machined voices 

to read books aloud, Google Translate for voicing out loud translations with precise 

pronunciation and human-like intonation, and many other applications which command 

ascended Artificial Intelligence (AI) technologies. 

WaveNet, a software developed by Google's UK-based AI firm DeepMind, belonging to 

Google since 2014 due to Google's acquisition, generates Google Cloud Text-to-Speech. The 

system makes an effort to set itself apart from Amazon and Microsoft. DeepMind's AI speech 

synthesis technology is notable for its realistically sophisticated naturalness. Concatenative 

synthesis is used by most speech synthesizers (including one of the most powerful AI apps - 

Apple's Siri), where a program saves individual phonemes and subsequently puts them together 

to produce words, phrases, or sentences. WaveNet, on the other hand, generates speech via 

machine learning. It then reconstructs waveforms from a database of the human voice with a 

rate of up to 24,000 samples a second. Google Text-to-Speech has successfully exploited new 

technology - speech synthesis- to the full. 

A variety of voices and languages supports GTTS. 

GTTS is provided with a wide range of voices, including both standard and WaveNet voices. 

WaveNet voices, specifically, are examined for higher quality voices. 

 



ACOJ- ISSN 1936-9859 AsiaCALL Online Journal  Vol. 13; No. 1; 2022 

97 

 

Particularly, WaveNet voice implemented in Google Translate - Google Text-to-Speech is coded: 

Language  Voice type Language code Voice name SSML Gender 

English (US) WaveNet en-US en-US-Wavenet-G FEMALE 

GTTS is believed to be the highest achiever on the Mean Opinion Score (MOS) scale. 

This bar chart below indicates that WaveNet voices in US English and Mandarin Chinese are 

significantly more outweighed compared to other synthetic voices and human speech. The y-

axis values show raters’ Mean Opinion Score (MOS). On a scale of 1 to 5, test raters evaluated 

each voice based on how much it was almost like natural human talk. (Google, 2013) 



https://asiacall.info/acoj Le, P. N., Vu, H. M. L., & Tran, M. N. Vol. 13; No. 1; 2022 

98 
 

 

Figure 1.  Comparison between US English and Mandarin Chinese using WaveNet voices, other 

speech synthesis and human speech.   

WaveNet produces more natural-sounding voices than previous text-to-speech model systems. 

Its synthesized voices are based upon levels of phonemes, syllables, and words which shadow 

human-like emphasis and inflection. In 2016, Oord, Dieleman, Zen, Simonyan, Vinyals, 

Graves., ... & Kavukcuoglu, K. proved the effectiveness and the degree of naturalness of 

WaveNet voices via their paper. The researchers conducted three experiments in total, including 

Multi-Speaker Speech Generation (MSSG), Text-to-Speech (TTS), and Music (M) in order to 

give clear evidence and significant statistics. Among others, it is rather obvious to view from 

Table 1 that TTS from WaveNet occurred to be greater outperformed the two models which 

were first from Gonzalvo et al. (2016) - hidden Markov model (HMM) and the other was from 

Zen et al. (2016) - long short-term memory recurrent neural network (LSTM-RNN). WaveNet 

voices reached 4.21 for North American English and 4.08 for Mandarin Chinese on the 5-point 

MOS scale for its naturalness, which proved the undeniable significance. They were considered 

the surprisingly highest point ever achieved. WaveNet considerably improved the prior state of 

the art, narrowing down the distance between human-like speech and the best previous 

synthesized speech model by up to 50%.  

Table 1 The MOS test results are shown from the test.  

 

2.1.3. Shadow-reading technique 

Intonation should be substantially put into consideration to teach at the discourse genres, 

according to Levis (2002; 2004) and Jenkins (2004). Shadowing, along with mirroring, 



ACOJ- ISSN 1936-9859 AsiaCALL Online Journal  Vol. 13; No. 1; 2022 

99 

repetition, and imitation approaches (Goodwin 2004), is one of the most effective oral teaching 

methodologies applied for copying intonation of native speakers, according to Celce-Murcia et 

al. (1996). In practical application, the shadowing technique is commonly employed in 

Simultaneous Interpretation training (SI). Before starting SI training sections, the participants 

must engage in extensive shadowing practices to grasp the rhythm Shadowing Technique in 

English Intonation Instruction and prosodic elements in actual speech. The fundamental skill 

set of shadowing is to closely follow the utterances of Native Speakers (NSs) (Luo, Yamauchi 

& Minematsu, 2010). 

In this paper, a pedagogical method called shadow-reading technique, adapted from Murphey's 

(2001a) - deemed to be 'conversational shadowing', predominantly became our central 

instruction. The pedagogical method helps to strike into a complicated yet customary 

phenomenon in discourse matters regarding people's silent or aloud repetition of what they hear 

from their interlocutors. Shadowing, from 1992, has been identified as an effective tool for 

training interpreters (Kurz, 1992; Sabatini, 2000) and for assessing skills in the language 

process, particularly for listening and reading skills, to majority scholars (Kurz, 1992; Sabatini, 

2000). (Gray, 1975; Jordan, 1988; Marslen-Wilson, 1975; Muchisky, 1983). 

Murphey (2001a) advocated for the use of conversational shadowing in classroom research, a 

practice in which students not only repeat, partially or totally, what other students say but also 

respond dialogically. Summarizing as a means of shadowing lengthy spans of conversation is 

also recommended by Murphey (2001a). Therefore, there are two formations of shadowing: 

shadowing and summarizing. According to Murphey (1996, 2001a), Shadowing may be useful 

in demonstrating to learners how to make negotiations in a way that facilitates L2. As indicated 

by Gal'perin (1969) approach to the development of mental processes, interactive shadowing 

and summarizing appear to be excellent for external (audio) verbalization. Shadowing and 

summarizing have been demonstrated to be useful in boosting L2 learners' listening and 

speaking abilities in empirical studies (Tabata, 2002; Tamai, 2001). With that being said, 

shadowing itself became our exclusive method option to implement in our online training. 

Vygotsky's (1986) idea of imitation has been shown to be in agreement with James M. 

Baldwin's (Lantolf & Thorne, 2006), particularly the form of imitation he called persistent 

imitation. Simple imitation, which Baldwin (1894, 1895/1906) defined as an involuntary, 

results in a more or less close copy of the original and does not involve consciousness of what 

has been done, and persistent imitation, on the other hand, is a type of purposeful, intentionally 

cognitive activity that involves making multiple tries to duplicate a model, each time in an 

attempt to improve the reproduction using an image of the original model and the imitator's past 

attempts as a guide. Therefore, persistent imitation is more cyclically, purposefully 

transformative in nature and optioned as a more narrowed scope of the study. 

It is apparent that imitation does not imply mindless mimicry or repetition. Imitation entails 

recognizing the means to achieve the activity's goal (end, purpose, or result) and understanding 

the objective's aim (end, purpose, or result). Linguistic imitation, in particular, necessitates both 

awareness of the communication goals and the tendency to generate language. Rather, it is the 

endeavor of children to recreate the language that adults produce with the same communication 

function, according to Tomasello (2002). Two important factors govern linguistic imitation: 

entrenchment and pre-emption (Tomasello, 2003). Learners get habituated to producing 

language solely in ways that are comparable to those generated by professional speakers around 



https://asiacall.info/acoj Le, P. N., Vu, H. M. L., & Tran, M. N. Vol. 13; No. 1; 2022 

100 
 

them through entrenchment, which is achieved by frequent exposure to linguistic models and 

successful use. 

2.2. Previous studies 

There is increasing demand in learning global languages in this interconnected world, and 

English has long taken the lead amongst other languages. Language, by far, is one of the primary 

factors that function as a bridge to bring worldwide nations closer, which means a great deal in 

terms of both economic and social aspects (Tran & Nguyen, 2018, as cited in Luu et al., 2021). 

As far as language acquisition is concerned, EFL learners are expected to master all four skills 

to be verified as proficient language users. Henceforth, the standardized English language 

syllabus used in Vietnam has long been dominated by the teaching of grammar and reading 

comprehension while the speaking skill was hardly paid any attention to, which resulted in 

learners not being able to communicate in a real-time context properly. Pronunciation and 

intonation were therefore also excluded as part of the national syllabus for the foreign language 

program for ages before they started to captivate attention from Vietnamese researchers and 

educators roughly two decades ago. Since then, with the advancement of technologies, and 

recently, the outbreak of the Covid-19 pandemic, learners all over the world have been 

subjected to virtual learning. Though met with a barrage of criticism, online learning, assisted 

by up-to-date media, presents a “strong potential to enhance language skills for learners and 

promote the process of learning English" (2021, Van et al., as cited in Pham, 2021), and a vast 

variety of research into this particular area has got an opportunity to expand further. 

In 2011, Ngo published a study featuring adopting an intonation teaching approach that 

involved individual language learning strategies (LLS). The training model allowed learners to 

decide on the strategy that best suited them, and it was shown to work well in combination with 

other teaching and learning techniques employed throughout the training course, such as written 

reports, field notes, group discussions and interviews. As a result, the study generally revealed 

improvement in the ability of Vietnamese learners of English to perceive two features of 

English intonation: tonicity and tone after the training course. Within the same year, Tran (2011) 

also suggested that the teaching of intonation in which students are entitled to experiencing real 

English rhythmic patterns and melody yielded a better result on students’ performance of 

intonation compared to those who were only provided with theoretical knowledge concerning 

intonation. Most of these studies, despite particular differences in their approaches, revealed a 

positive outcome for certain implications on teaching and acquiring pronunciation/ intonation. 

Students’ interest was successfully aroused as they were allowed to have a more hands-on 

experience in their training and really get engaged in the world of melody with diverse prosodic 

features, which proves absolutely crucial in any pronunciation training. However, regardless of 

the observable merits, it goes without saying that those teaching methods not only take a large 

amount of time but also demand close monitoring of the trainer. Thus, whether it would present 

a mismatch between what the research team thinks is possible and what can actually be achieved 

after an intonation training course remains questionable. 

Studies on shadowing technique & intonation 

In the era of the Audio-Lingual method, shadowing was considered one of the most relevant 

techniques to use in teaching speaking and listening, and so far, it still proves to be an effective 

tool in pronunciation training (Лобачова, 2020). In fact, practically speaking, of all the 

techniques prevalently implemented in intonation training, shadowing is gaining greater 



ACOJ- ISSN 1936-9859 AsiaCALL Online Journal  Vol. 13; No. 1; 2022 

101 

support than ever and contributing considerably to the second language teaching as well as 

acquisition. However, it is rather surprising that back in the early 2000s, little research could be 

observed on the implementation of shadowing in teaching speaking, with pronunciation and 

intonation included (Hamada, 2018). Despite its own failure to maintain researchers' interest 

over the last two decades, shadowing technique is currently intensively delved into and 

manifests itself in an abundance of research schemes into the field of intonation training 

worldwide. Up to the present, most of the works aiming at measuring the effectiveness of 

teaching experiments using shadowing technique have revealed a positive impact on students' 

performance in receptive and productive skills such as listening, speaking and reading (Hamada, 

2016; Nakanishi & Ueda, 2011; Hsieh, Dong, & Wang, 2013). The prosperity of research into 

the importance of shadowing for English language teaching is rather obvious, but then, there 

are few studies that examine the influence of this particular technique on pronunciation in 

general and intonation to be more specific. In addition, despite the fact that different training 

approaches were taken into detailed consideration and all proved fruitful to some extent, they 

tend to involve great effort and require teachers to have a profound understanding of prosodic 

features and intonation-related concepts. Nonetheless, it is not ideal in Vietnam, especially 

when most teachers lack the confidence to conduct pronunciation/intonation lessons. 

2.3. The current study 

The teaching of intonation has been neglected for a long time for numerous reasons (Ngo, 2011). 

Research by Graham & Post (2018) suggested that together with other aspects, the L2 

competency of a second language learner was also represented in the way they deal with 

intonation while speaking. However, in spite of its importance, there has been no course 

devoted to teaching intonation at any language institutions in Vietnam, along with the fact that 

the amount of time allocated for teaching intonation in the course syllabus is not specified, and 

even if any, the amount remains minimal and often skipped by the teachers as intonation 

training section is, more often than not, considered optional. This is true for all levels of public 

education in Vietnam, from primary schools to tertiary institutions, and the significance of 

intonation training seems to be overlooked at present time though several reformations 

regarding second language teaching have already taken place throughout the course of history. 

As intonation is not presented as part of the national syllabus of English subjects, it is not legally 

included in any kinds of testing or assessment of English language proficiency issued by the 

Vietnamese Ministry of Education and Training. This necessarily presents an urge to rethink 

the problem, raising teachers' awareness of the fact that intonation training should not be 

undervalued under any circumstances and coming up with in-time strategies to compensate for 

the ever-gloomy scenario of intonation teaching in Vietnam. 

The lack of research into intonation training that takes into account the Vietnamese educational 

settings could be attributed to numerous factors, among which stands out two significant ones 

that, as the research team believes, can be properly addressed within the scope of this study. 

First, the lack of confidence and understanding of intonation amongst Vietnamese teachers of 

English remains quite a prevalent issue (Ngo & Setter, 2011). Second, intonation, like other 

language areas, requires learners to have constant exposure to it by listening to modal 

recordings as well as practicing using appropriate intonation on a regular basis in order to be 

familiarized with intonation. Also, assessing students’ progress in their performance can even 

present itself as a major concern for those in charge of teaching intonation. The question pops 

up right here: in a circumstance in which intonation training is generally undervalued, 



https://asiacall.info/acoj Le, P. N., Vu, H. M. L., & Tran, M. N. Vol. 13; No. 1; 2022 

102 
 

represented by the fact that no course has ever been specifically designed for reinforcing 

students' perception of intonation, and economic factors constantly prevent a large number of 

Vietnamese learners from accessing proper training, what should be done to address the 

problem? At this point, it is worth remembering that the world has entered the greatest digital 

reformation, and the options for any kind of learning should be numerous. Zeitoun (2008, as 

cited in Nguyen, 2021), classified online learning into two different types: synchronous and 

asynchronous mode. The former is typically conducted via a video-conferencing room, which 

allows learners to interact intensively with their instructors and receive different kinds of 

feedback, either impromptu or official, while the latter offers students access to a wide range 

of sources with already prepared lessons, and they can keep track of their own learning pace. 

Regarding intonation training, the advent of applications assisting the acquisition of 

pronunciation, ELSA speak, Voki, and Puppet Pals, to name but a few, has proved its great 

contribution to improving the situation (Samad and Ismail, 2020). However, most of them are 

not free of charge or have a free basic version which merely provides learners with limited 

access to their features and resources. By now, the option should be clear – the research team 

is seeking a tool or website assisting intonation practice that (1) possesses a wide range of 

resources of oral records and database to serve the purposes, (2) fully requires little competency 

from the teachers and (3), most importantly, easily accessed with permanent free of charge 

policy. After researching intensively, GTTS seemed to be a potential candidate and was taken 

into account as the primary tool adopted in this experimental study. Specifically, GTTS operates 

on a massive volume of worldwide language data, with prosody information (i.e., rhythm and 

melody) included, and the voice assistant embedded in the website is totally eligible for a 

pronunciation/intonation learning tool. Besides, GTTS, which is embedded within Google 

Translate, is a cost-free and readily available service, with a new model introduced recently that 

enables users to use the app everywhere, no matter whether there is an internet connection or 

not ("About - Google Translate", 2017). These features make GTTS a preferred tool for our 

intonation training scheme. Basically, the study aims at measuring the effects of the shadowing 

technique aided by GTTS on EFL students' performance of intonation in-text throughout a pre-

designed training course. 

2.4. Research questions 

The aims of this study can be addressed through the following research questions: 

1. Do the students use default intonation in text correctly before the online training course? 

2. To what extent do the students improve their performance of default intonation in text after 

the online training course? 

 

3. Methods 

3.1. Pedagogical setting and participants 

The participants in the study were university students ranging from first to the third year who 

have little to no knowledge of English intonation in text. At the time of the research, these 

students were studying at two universities in Ho Chi Minh City (abbreviated to HCMC), 

Vietnam, namely the HCMC University of Technology and the HCMC University of 

Education. Both convenience sampling and snowball sampling were employed for the 

distribution of the questionnaire. First, the research team posted an online questionnaire via 



ACOJ- ISSN 1936-9859 AsiaCALL Online Journal  Vol. 13; No. 1; 2022 

103 

Google Forms on the Facebook pages of the two universities mentioned previously, which 

received a total of 172 responses. The respondents, then, were kindly requested to invite their 

classmates and friends to fill in the form. Hence a snowball sampling started, which gathered 

ten more participants in the process. The responses were then checked manually to opt out of 

those who had more competency in English intonation needed for the study. In total, 18 valid 

responses were recorded. Due to the adversity of the raging Covid-19 pandemic, the size of the 

participants was then shrunk to only 7. This represented a great difficulty for the research team 

in finding a sufficient number of available participants to take part in the online training course. 

Despite the very limited sample given, the research team believes that this study's findings 

should provide a starting point for further in-depth research on the same area of interest. 

3.2. Design of the study 

The study was conducted using both quantitative and qualitative components. A pretest/posttest 

design was used to measure the effects of a 10-week online training course (ten 90-minute 

sessions on the Zoom online meeting platform) on English default intonation. The online course 

only took place in only 10 weeks since it occurred during the start of the participants’ new 

school year. Periods longer than 10 weeks would have been problematic for the participants’ 

schedules. The course syllabus mainly focused on the default intonation of five types of 

sentences, namely Statement, Exclamation, Yes-no question, Wh-question, and Tag question. 

Although the course included all types of default intonation in text, the research team only 

elaborated on falling and rising intonation as the study's aim. The course allowed the saving of 

the audio files, which facilitated the contrast between the prosodic patterns produced by the 

participants and that of the spoken texts provided by GTTS. These files were then analyzed 

using Speech Analyzer software to provide auditory and visual displays of pitch contours as 

feedback and assessment. The software, hence, served as a tool for the interpretation of the 

study's findings. The training course syllabus followed a framework of shadowing instruction 

and structured input activities (Wells, 2006) in combination with GTTS. The input activities 

focused on helping the participants to recognize and reproduce some important patterns of 

English default intonation in text. The course also included explicit information about the 

patterns and their meaning in terms of text and basic conversation (Wells, 2006). 

3.3. Data collection and analysis 

The pretest (see Appendix 1) consisted of the first recording of the participants’ short sentences 

prior to receiving any target knowledge of English default intonation in text. The test required 

the students to create separate recordings of 6 sentences, including 2 statements, 1 exclamation, 

1 yes-no question, 1 wh-question, and 1 tag question. The results taken from the pretest will 

facilitate the course syllabus as they will provide the research team with a preliminary 

assessment of the participants' appropriate use of English default intonation in text. The 

researcher compiled notes on the sessions and the learners' comments and reactions during the 

training course. The participants' progress was also recorded for analysis and feedback 

throughout the course. At the end of the online training course, the posttest consisted of the 

participants' perception of default intonation in the text as well as the last version of short 

sentences produced and recorded by them. In terms of the recording questions, the students 

were asked to create separate recordings of 9 sentences, including 2 statements, 1 exclamation, 

3 yes-no questions, 1 wh-question, and 2 tag questions. Each participant completed an 

anonymous questionnaire on the value of the online intonation training course received within 

the post-test. 



https://asiacall.info/acoj Le, P. N., Vu, H. M. L., & Tran, M. N. Vol. 13; No. 1; 2022 

104 
 

Finally, using the Speech Analyzer software, each of the participants' recordings in the posttest 

was assessed by comparing their prosodic patterns with those produced by GTTS. Each of the 

recordings was assessed by the three members of the research team to make sure that the 

assessment was done coherently and appropriately. Then, each member gave their own 

comments on the degree of intelligibility of the participants' prosodic performance. After that, 

the research team provided feedback for the participants' performance. 

 

4. Results, Findings and discussion 

4.1. Results and interpretation 

As previously mentioned, participants were asked to do a pre-designed test on their ability to 

use proper intonation for different types of English sentences prior to the training course. It is 

rather surprising to observe that the students managed to use appropriate falling intonation 

while the rising tone seemed to pose a real challenge to their perception (1). 

After the training course, students' performance in intonation was assessed via another test 

whose results then, indicated an improvement in rising intonation. By contrast, it is noticed that 

their use of falling tone hardly improved. This trend can be presented in the charts as below: 

STATEMENT (FALLING INTONATION AS DEFAULT) 

Table 4.1 Frequency description of students’ pretest default falling intonation performance 

using Speech Analyzer 

PRETEST 

Sentence Name 
Nucleus 

(Hz) 

Tail 

(Hz) 

Percentage 

(%) 

1. I don't think it is true 

that she was afraid. 

Student A 287 211 73.52 

Student B 307 254 82.74 

Student C 118.7 105.9 89.22 

Student D 140.9 66.1 46.91 

Student E 419 243 58.00 

Student F 136 120 88.24 

Student G 294 184 62.59 

Google 296 113.4 38.31 

 

 

 



ACOJ- ISSN 1936-9859 AsiaCALL Online Journal  Vol. 13; No. 1; 2022 

105 

Table 4.2 Frequency description of students’ posttest default falling intonation performance 

using Speech Analyzer 

POSTTEST 

Sentence Name 
Nucleus 

(Hz) 

Tail 

(Hz) 

Percentage 

(%) 

5. I believe it is going to 

rain at the 

end of the week. 

Student A 271 210 77.49 

Student B 246 195 79.27 

Student C 139.7 119.8 85.76 

Student D 137.8 122.8 89.11 

Student E 294 214 72.79 

Student F 375 68.8 18.35 

Student G 280 176 62.86 

Google 297 122.8 41.35 

The tail movement percentage illustrates whether the sentence is uttered with falling or rising 

intonation. A value below 100% points to a fall in intonation, while one exceeding 100% 

indicates a rise in the speaker's speech. 

The table is intended to record the figures for statements. Obviously, tail movement percentage 

values generated by the speech of all students manifest themselves below 100%, so all those 

taking the test used proper falling intonation for this sentence type in both pretest and posttest. 

That means the learners generally did not struggle with the falling intonation, and there is also 

no significant improvement in learners' perception of this prosodic feature. 

YES-NO QUESTIONS (RISING INTONATION AS DEFAULT) 

 

 

 

 

 

 

 

 

 



https://asiacall.info/acoj Le, P. N., Vu, H. M. L., & Tran, M. N. Vol. 13; No. 1; 2022 

106 
 

Table 4.3 Frequency description of students’ pretest default rising intonation performance 

using Speech Analyzer 

PRETEST 

Sentence Name 
Nucleus 

(Hz) 

Tail 

(Hz) 

Percentage 

(%) 

2. Is it true that you 

faked your way in 

here? 

Student A 344 283 82.27 

Student B 357 287 80.39 

Student C 111 106.4 95.86 

Student D 64.9 229 352.85 

Student E 247 457 185.02 

Student F 160 139 86.88 

Student G 406 234 57.64 

Google 237 286 120.68 

 

Table 4.4 Frequency description of students’ posttest default rising intonation performance 

using Speech Analyzer 

POSTTEST 

Sentence Name 
Nucleus 

(Hz) 

Tail 

(Hz) 

Percentage 

(%) 

1. Are you serious? 

Student A 331 355 107.25 

Student B 246 260 105.69 

Student C 129.9 231 177.83 

Student D 127.8 227 177.62 

Student E 237 376 158.65 

Student F 146.9 185 125.94 

Student G 227 228 100.44 

Google 267 287 107.49 



ACOJ- ISSN 1936-9859 AsiaCALL Online Journal  Vol. 13; No. 1; 2022 

107 

The research team attempted to calculate the variation percentage concerning Yes-No questions 

brought about by both the pretest and the posttest. According to Wells (2006), the default 

intonation of Yes-No question is a rise. 

As to the pretest, most figures are below 100%, which points to a fall in the participants’ speech. 

This is deemed inappropriate as the participants were required to record the yes-no question 

given using proper default intonation, i.e., a rise. Meanwhile, the above-100% percentages 

shown in the posttest suggest that the learners successfully used the correct rising intonation as 

default. 

At this point, it would be better to delve into the dilemma in greater depth. The fact that learners 

barely improve their intonation when it comes to a fall can be attributed to two main reasons. 

First, Vietnamese intonation does not embed and operate on semantic meaning but lexical 

meaning, which features the rise or fall based merely on the tonal systems with the combination 

of 6 fundamental tones. That should explain why Vietnamese do not have a tendency to lower 

their tone when asserting discourse-level definiteness and certainty. Secondly, according to 

Wichmann (2014), when learners are trained properly and their awareness is deliberately 

activated, they tend to improve rising more than falling intonation. Theoretically, the rising tone 

is easier to recognize. 

To exemplify (1), students’ intonation performance in 3 sentences corresponding to two types 

of intonation - rising and falling - will be placed under thorough scrutiny. The measurement of 

improvement follows the rating criteria below (table 4.5): 

 

Table 4.5 Improvement Scale 

Score Rating Definition 

1 Worse No improvement 

2 Mildly improved < 25% improvement 

3 Improved 25-49% improvement 

4 Much improved 50-74% improvement 

The table above is adapted from the Subject Global Aesthetic Improvement Scale (2019). It is 

used to measure and interpret the difference between the mean baseline and mean follow-up 

grades of the participants’ prosodic patterns. 



https://asiacall.info/acoj Le, P. N., Vu, H. M. L., & Tran, M. N. Vol. 13; No. 1; 2022 

108 
 

 

Figure 4.6 Tail movement description 

 

Improvement (%) = |(LPost - GPost) - (LPre - GPre)|  

Note: L = Learner’s tail movement percentage in specific sentence type 

          G = GTTS tail movement percentage in specific sentence type 

(LPost - GPost) = percentage of the difference between learner's performance of 

falling/rising tones and GTTS' performance on falling/rising tones in POSTTEST 

(LPre - GPre) = percentage of the difference between learner's performance of 

falling/rising tones and GTTS' performance on falling/rising tones in PRETEST 

This formula is applied to the analysis of the participants’ improvement in all types of sentences, 

which will be elaborated as follows: 

● Default falling intonation 1 (Statement) 

 

 

 

 

 

 

 

 



ACOJ- ISSN 1936-9859 AsiaCALL Online Journal  Vol. 13; No. 1; 2022 

109 

Table 4.7 The improvement of the participants’ default falling intonation 

Sentence type Name 

PRETEST POSTTEST 

Improvement 

(%) Tail movement 

percentage (%) 

Tail movement 

percentage (%) 

Statement 

Student A 73.52 77.49 0.94 

Student B 82.74 79.27 6.50 

Student C 89.22 85.76 6.50 

Student D 46.91 89.11 39.17 

Student E 58.00 72.79 11.76 

Student F 88.24 18.35 72.92 

Student G 62.59 62.86 2.76 

Google 38.31 41.35  

 

Figure 4.7a Student E’s PRETEST performance of a Statement 



https://asiacall.info/acoj Le, P. N., Vu, H. M. L., & Tran, M. N. Vol. 13; No. 1; 2022 

110 
 

 

Figure 4.7b Student E’s POSTTEST performance of a Statement 

As depicted by Table 4.7, Figures 4.7a and 4.7b, there is a slight overall improvement 

regarding the participants' default falling intonation in statements. Specifically, while one 

student shows a significant improvement of 72.92% in their default falling intonation and one 

with a considerable improvement of 39.17%, four others show mild improvement of more than 

2%. However, one of them does not show any improvement after the online training course. 

● Default falling intonation 2 (Exclamation) 

Table 4.8 The improvement of the participants’ default falling intonation 

Sentence type Name 

PRETEST POSTTEST 
Improvement 

(%) Tail movement 

percentage (%) 

Tail movement 

percentage (%) 

Exclamation 

Student A 64.14 63.29 5.28 

Student B 149.43 24.93 128.93 

Student C 76.95 32.49 48.89 

Student D 62.25 63.27 3.41 

Student E 64.33 62.21 6.55 

Student F 105.36 29.93 79.86 

Student G 34.88 60.59 21.28 

Google 48.33 52.76  



ACOJ- ISSN 1936-9859 AsiaCALL Online Journal  Vol. 13; No. 1; 2022 

111 

 

Figure 4.8a Student B’s PRETEST performance of an Exclamation 

 

Figure 4.8b Student B’s POSTTEST performance of an Exclamation 

Table 4.8, Figures 4.8a and 4.8b illustrate an overall mild improvement in the students’ default 

falling intonation in exclamations. Two students among the group perform significantly better 

with 128.93% and 79.86% in the improvement rate. One of which shows a mild improvement 

of 48.89% in their English default falling intonation. The remaining four students show a slight 

improvement of more than 3% in their intonation performance. 

● Default rising intonation 2 (Yes-No question) 

 

 

 

 

 

 

 

 

 

 

 

 



https://asiacall.info/acoj Le, P. N., Vu, H. M. L., & Tran, M. N. Vol. 13; No. 1; 2022 

112 
 

Table 4.9 The improvement of the participants’ default rising intonation 

Sentence type Name 

PRETEST POSTTEST 
Improvement 

(%) Tail movement 

percentage (%) 

Tail movement 

percentage (%) 

Yes-No 

question 

Student A 82.27 107.25 38.17 

Student B 80.39 105.69 38.48 

Student C 95.86 177.83 95.16 

Student D 352.85 177.62 162.04 

Student E 185.02 158.65 13.19 

Student F 86.88 125.94 52.25 

Student G 57.64 100.44 55.99 

Google 120.68 107.49  

 

Figure 4.9a Student A’s PRETEST performance of a Yes-No question 



ACOJ- ISSN 1936-9859 AsiaCALL Online Journal  Vol. 13; No. 1; 2022 

113 

 

Figure 4.9b Student A’s POSTTEST performance of a Yes-No question 

According to Table 4.9, Figures 4.9a and 4.9b, the students show an overall significant 

improvement regarding the participants’ default rising intonation in Yes-no questions. In 

particular, one student performs 162.04% better in the posttest than in the pretest, which is a 

significant improvement. Three other students show considerable improvement of more than 

50%, while the three remaining show mild improvement of more than 10%. 

Addressing the first research questions posed from the beginning of this paper, it is evident that, 

in the pretest, the students performed relatively well in sentences that required falling intonation 

while they seemed to struggle with rising intonation. For the second question, the results suggest 

that the participants made significant progress in their use of rising intonation, whereas their 

performance in falling intonation hardly improved after the training course. 

4.2. Merits of the study 

Despite only a small number of learners taking part in the study, the research team is able to 

demonstrate a relatively positive result obtained by an alternative to costly tools and time-

consuming approaches to teaching intonation. With the tool supported by Google and the 

framework clearly presented in the coursebook, it is reasonable to expect that teachers will be 

less likely to suffer from hard work and the fear of their own ignorance of the prosodic word. 

In addition, with the aid of the widely-available Google Text-to-Speech system, learning does 

not necessarily occur at school, which addresses a long-standing problem that no specific time 

is assigned to intonation teaching at school currently. Last but not least, the idea of taking 

Google Text-to-Speech as a model to imitate provides students with a purpose to learn. Indeed, 

85% of those participating in the project reported that they felt a sense of achievement as they 

managed to imitate exactly what the machine said.  

As claimed in the first place, the study delves into a new approach to intonation training, which 

is believed to benefit EFL teachers and learners in various ways. The vast availability of GTTS 

grants teachers easy access to one of the most highly approved AI applications in the modern 

world, with a huge authentic linguistic database. There is little syllabus time as well as few 

financial challenges involved, as the practice of intonation is not only refrained in the classroom 

setting, and users are not required to make registration or process any problematic installation. 

This study also successfully reveals a positive influence of the combination of GTTS and 

shadowing technique on learners' performance in intonation, especially when it comes to rising 

tone. 

 



https://asiacall.info/acoj Le, P. N., Vu, H. M. L., & Tran, M. N. Vol. 13; No. 1; 2022 

114 
 

5. Conclusion 

All in all, although the study succeeds in yielding certain positive results in learners' use of 

intonation, the small participant size, to a lesser or greater extent, prevents us from reaching 

any level of generalization for a larger population. Additionally, while the research team finds 

themselves in a rather fine position to answer the question of whether there is any learners' 

improvement in terms of intonation use, it presents, so far, an unsolved problem regarding the 

question of to what extent. This necessitates further research into the approach to gain more 

insight into its possible effects on intonation training and learners' perception of related prosodic 

features. 

Apart from these, it should also be mentioned that the scope of this study does not seem to 

afford generalization of the result as it is rather small. Right from the onset, the research team 

encountered a number of difficulties in recruiting the participants. Although the initial 

questionnaires managed to reach out more than 100 participants, with approximately 70% 

showing their support when asked if they would like to participate in an online intonation 

training, only 14 showed up on the first day and the number then dropped down to only 7 

towards the end of the training section. This can be attributed to 3 fundamental reasons. First, 

there is no commitment amongst those taking part in the project, though they were asked to 

affirm their commitment via a so-called agreement between the learners and the researchers. 

The fact that no punishment was administered for their one-sided drop might result in the sense 

of little obligation to the course. Without obvious gain and loss, it is virtually impractical to 

believe they would adhere to the training as, generally, students are not intrinsically motivated. 

Second, the course was initiated at the start of an academic year when university students 

seemed to be fully occupied in preparing for their classwork. Getting engaged in a course with 

an unforeseeable outcome could prove a waste of time to those already packed with their daily 

schedule. Another reason is that online learning, despite its particular merits, might seem 

daunting to a number of learners. 

 

References 

"About - Google Translate". Google Translate. Retrieved July 9, 2017.  

"Google Text-to-Speech - Apps on Google Play". play.google.com. 

Anh, N. P. (2017). Teaching English intonation to Vietnamese students of English: Tonicity 

and tone. VNU Journal of Foreign Studies, 33(6). 

Baldwin, J.M. (1895/1906). The mental development of the child and the race (3rd ed.). New 

York: Macmillan. Retrieved from: 

http://www.brocku.ca/MeadProject/Baldwin/Baldwin_1906/Baldwin_1906_toc.html 

(May 2013). 

Cruttenden, A. (1997) Intonation (2nd edn). Cambridge: Cambridge University Press. 

Crystal, D. (1986) Prosodic development. In P. Fletcher and M. Garman (eds) Studies in First 

Language Development (pp. 174–97). New York: Cambridge University Press. 

Gal’perin, P.Y. (1969). Stages in the development of mental acts. In M. Cole, & I. Maltzman 

(Eds.), A handbook of contemporary Soviet psychology (pp. 249–273). New York: Basic 

Books. 

http://www.brocku.ca/MeadProject/Baldwin/Baldwin_1906/Baldwin_1906_toc.html


ACOJ- ISSN 1936-9859 AsiaCALL Online Journal  Vol. 13; No. 1; 2022 

115 

Gilbert, J. (1994). Intonation: A navigation guide for the listener. Pronunciation pedagogy and 

theory: New views, new directions (ed. by Joan Morley), 36-48. Alexandria, VA: 

TESOL. 

Gonzalvo, Xavi, Tazari, Siamak, Chan, Chun-an, Becker, Markus, Gutkin, Alexander, and 

Silen, Hanna. Recent advances in Google real-time HMM-driven unit selection 

synthesizer. In Interspeech, 2016. URL 

http://research.google.com/pubs/pub45564.html. 

Graham, C., & Post, B. (2018). Second language acquisition of intonation: Peak alignment in 

American English. Journal of phonetics, 66, 1-14. 

Gray, M.J. (1975). Effects of shadowing on reading. Journal of Experimental Psychology: 

Human Learning and Memory, 104, 423–428. 

Halliday, M.A.K. (1970) A Course in Spoken English: Intonation. London: Oxford University 

Press. 

Halliday, M.A.K. (1994) An Introduction to Functional Grammar. London: Arnold. 

Hamada, Y. (2016). Shadowing: Who benefits and how? Uncovering a booming EFL teaching 

technique for listening comprehension. Language Teaching Research, 20(1), 35-52. 

Hamada, Y. (2018). Shadowing for pronunciation development: Haptic-shadowing and IPA-

shadowing. Journal of Asia TEFL, 15(1), 167. 

Hsieh, K. T., Dong, D. H., & Wang, L. Y. (2013). A preliminary study of applying shadowing 

technique to English intonation instruction. Taiwan Journal of Linguistics, 11(2). 

http://ebookbrowse.com/teaching-pronunciation-in-the-leamer-centered-classpdf-

dl01773540 

Jenkins, J. (2004). Research in teaching pronunciation and intonation. Annual Review of 

Applied Linguistics 24: 109-125. 

Jordan, N.C. (1988). Language processing and reading ability in children: A study based on 

speech-shadowing techniques. Journal of Psycholinguistic Research, 17, 357–377. 

Kurz, I. (1992). ‘Shadowing’ exercises in interpreter training. In C. Dollerup, & A. 

Loddengaard (Eds.), Teaching translation and interpreting: Training, talent and 

experience (pp. 245–250). Amsterdam: John Benjamins. 

Lantolf, J.P., & Thorne, S.L. (2006). The sociogenesis of second language development. 

Oxford: Oxford University Press. 

Levis, J. M. (2002). Reconsidering low-rising intonation in American English. Applied 

Linguistics U.V. 56-82. 

Levis, J. M. and Pickering, L. (2004). Teaching intonation in discourse using speech 

visualization technology. System 32.4: 505-524. 

Lieberman, P. (1986) The acquisition of intonation by infants: Physiology and neural control. 

In C. Johns-Lewis (ed.) Intonation in Discourse (pp. 239–57). London: Croom Helm. 

Lin, Hsiang-Pao., Chuen-Yn Fan, and Chi-Fen Chen. 1995. Teaching pronunciation in the 

learner-centered classroom. Retrieved January 11, 2011 from 

http://research.google.com/pubs/pub45564.html.
http://research.google.com/pubs/pub45564.html.
http://research.google.com/pubs/pub45564.html.
http://ebookbrowse.com/teaching-pronunciation-in-the-leamer-centered-classpdf-dl01773540
http://ebookbrowse.com/teaching-pronunciation-in-the-leamer-centered-classpdf-dl01773540


https://asiacall.info/acoj Le, P. N., Vu, H. M. L., & Tran, M. N. Vol. 13; No. 1; 2022 

116 
 

Luo, D., Yutaka, Y., and Nobuaki, M. (2010). Speech analysis for automatic evaluation of 

shadowing. Paper presented at the Interspeech 2010 Joint Satellite Workshop on 

"Second Language Studies: Acquisition, Learning, Education and Technology", Tokyo: 

Japan. 

Luu, L. P. T., Nguyen, T. N. Q., Vo, N. T. T., & Nguyen, M. T. H. (2021). The Need of 

Applying English Learning Apps to Help Van Lang University Students Improve Their 

Spoken English Performance. AsiaCALL Online Journal, 12(2), 72-86. Retrieved from 

https://asiacall.info/acoj/index.php/journal/article/view/33 

Marslen-Wilson, W.D. (1975). Sentence perception as an interactive parallel process. Science, 

189, 226–228. 

Mehler, J., Jusczyk, P.W., Lambertz, G., Halsted, N., Bertoncini, J. and Amiel-Tison, C. (1988) 

A precursor of language acquisition in young infants. Cognition 29, 143–78. 

Muchisky, D.M. (1983). Relationships between speech and reading among second language 

learners. Language Learning, 33, 77–102. 

Murphey, T. (1996, September). Proactive adjusting to the zone of proximal development: 

Learner and teacher strategies. Paper presented at the 2nd Conference for Socio-

cultural Research Vygotsky–Piaget, Geneva, 1–4. 

Murphey, T. (2001a). Exploring conversational shadowing. Language Teaching Research, 5, 

128–155. 

Nakanishi, T., & Ueda, A. (2011). Extensive reading and the effect of shadowing. Reading in 

a Foreign Language, 23(1), 1-16. 

Ngo, A. P., & Setter, J. (2011, August). Evaluating the Intonation Perception Ability of 

Vietnamese Learners of English. In ICPhS (pp. 1466-1469). 

Nguyen, H. N. (2021). A Case Study of Home-based Teaching and Learning Practice at Ton 

Duc Thang University. AsiaCALL Online Journal, 12(2), 1-11. Retrieved from 

https://asiacall.info/acoj/index.php/journal/article/view/24 

Oord, A. V. D., Dieleman, S., Zen, H., Simonyan, K., Vinyals, O., Graves, A., ... & 

Kavukcuoglu, K. (2016). Wavenet: A generative model for raw audio. arXiv preprint 

arXiv:1609.03499. 

Pham, D. T. T. (2021). The effects of Audiovisual Media on Students’ Listening 

Skills. International Journal of TESOL & Education, 1(1), 13–21. Retrieved from 

http://i-jte.org/index.php/journal/article/view/3 

Pickering, L. (2004). The structure and function of intonational paragraphs in native and non-

native speaker instructional discourse. English for Specific Purposes 23A: 19-43. 

Sabatini, E. (2000). Listening comprehension, shadowing and simultaneous interpretation of 

two ‘non-standard’ English speeches. Interpreting, 5, 25–48. 

Samad, I. S., & Ismail, I. (2020). ELSA Speak Application as a Supporting Media in Enhancing 

Students’ Pronunciation Skill. Majesty Journal, 2(2), 1-7. 

Snow, D. and Balog, H. (2002) Do children produce the melody before the words? A review of 

developmental intonation research. Lingua 112, 1025–58. 

http://i-jte.org/index.php/journal/article/view/3


ACOJ- ISSN 1936-9859 AsiaCALL Online Journal  Vol. 13; No. 1; 2022 

117 

Supported voices and languages | Cloud Text-to-Speech Documentation 

Tabata, M. (2002). Shadowing and summarizing as effective output SLA activities. Nanzan 

Studies in English Language Education, 9, 115–170. 

Tamai, K. (2001). Strategic effect of shadowing on listening ability. In J. White (Ed.), 

Proceedings of the 4th international conference on foreign language education and 

technology (FLEAT IV) (pp. 620–625). Kobe: Japan Association for Language 

Education and Technology. 

Tomasello, M. (2002). The item-based nature of children’s early syntactic development. In M. 

Tomasello & E. Bates (Eds.), Language development: The essential readings (pp. 169–

186). Malden, MA: Blackwell. 

Tomasello, M. (2003). Constructing a language: A usage-based theory of language acquisition. 

Cambridge, MA: Harvard University Press. 

Vygotsky, L.S. (1986). Thought and language. Cambridge, MA: The MIT Press. 

Wells, J. C. (2006). English intonation PB and Audio CD: An introduction. Cambridge 

University Press. 

Wells, J. C. (2007). English Intonation. An Introduction. Cambridge University Press. 

Wennerstorm, A. (1998). Intonation as cohesion in academic discourse: A study of Chinese 

speakers of English. Studies in Second Language Acquisition 20.1: 1-25. 

Zen, Heiga, et al. (2016). Fast, compact, and high quality LSTM-RNN based statistical 

parametric speech synthesizers for mobile devices. In Interspeech, 2016. URL 

https://arxiv.org/abs/1606.06061 

Лобачова, І. (2020). Methodological aspects of implementing the shadowing technique in 

teaching english. Професіоналізм педагога: теоретичні й методичні аспекти, 

(13), 73-84. 

 

Biodata 

Nhi Le Phuong is currently studying for her Master’s degree in Linguistics at Hoa Sen 

University, HCMC. She fancies researching linguistic fields and teaching methodology. 

Linh Vu Hoang Mai is a visiting lecturer at Ton Duc Thang University and Hoa Sen University, 

Ho Chi Minh City, Vietnam. Currently she is doing a MA degree in Linguistics at Hoa Sen 

University and finds herself interested in the field of corpus linguistics. 

Nhat Tran Minh is a Testing coordinator at the Faculty of Foreign Languages at Hoa Sen 

University, Ho Chi Minh City, Vietnam. He is currently a Master candidate in Linguistics at 

Hoa Sen University who is interested in finding feasible teaching methods and applications in 

the 21st century. 

  

https://arxiv.org/abs/1606.06061
https://arxiv.org/abs/1606.06061
https://arxiv.org/abs/1606.06061


https://asiacall.info/acoj Le, P. N., Vu, H. M. L., & Tran, M. N. Vol. 13; No. 1; 2022 

118 
 

Appendix 

Appendix 1: PRETEST 

  Evaluation of English intonation performance   

  The following test is used to evaluate your English intonation performance as a 

means for the research team to design and adjust the online training course in the 

most effective way. 

As there are 6 sentences in the test, please prepare separate recordings for each 

one. Then, please upload them to the Google Forms accordingly (maximum 10MB 

per audio file). 

Note: 

- Please use the type of intonation that you have already known and refrain from 

referencing external sources for the test so that the research team may gather the 

most objective evaluation of your performance. 

- During recording, please make sure that your surrounding area is relatively quiet 

with minimum background noises. 

- After recording, please make sure that your audio file is clear enough for the 

listeners to understand. 

  

  Full name 

.......................................................... 

  

  Record the below sentences using the intonation that you have already known. 

1. I don't think it is true that she was afraid. 

2. Is it true that you faked your way in here? 

3. Which question is going to make them uncomfortable? 

4. Sounds impossible, doesn't it? 

5. What a great view! 

6. Don't rush or you will fall. 

Add file here 

 

  

 

 

 

 

 

 



ACOJ- ISSN 1936-9859 AsiaCALL Online Journal  Vol. 13; No. 1; 2022 

119 

Appendix 2: POSTTEST 

  Final Evaluation of English intonation performance   

  The following test is used to evaluate your final English intonation performance 

after taking the online training course as a means for the research team to evaluate 

the effectiveness of the method applied in the training course in improving 

students’ English default intonation performance. 

For the theoretical questions, please choose only ONE option or fill in the most 

appropriate answer that you have learnt. 

For the recording question, please prepare separate recordings for each of the 9 

given sentences. Then, please upload them to the Google Forms accordingly 

(maximum 10MB per audio file). 

Note: 

- Please use the type of intonation that you have already known and refrain from 

referencing external sources for the test so that the research team may gather the 

most objective evaluation of your performance. 

- During recording, please make sure that your surrounding area is relatively quiet 

with minimum background noises. 

- After recording, please make sure that your audio file is clear enough for the 

listeners to understand. 

  

  Full name 

.......................................................... 

  

  Theoretical questions 

Answer the following questions using what you have learnt (please refrain from 

using external sources during the test). 

Question 1: Intonation features: 

◯ the change of vocal pitch 

◯ the change of prosodic features 

◯ the rise of a person's voice to the highest point 

 Question 2: The use of monotone in reading/speaking indicates: 

◯ the overwhelming of emotions which bring many expressions 

◯ the lack of confidence 

  



https://asiacall.info/acoj Le, P. N., Vu, H. M. L., & Tran, M. N. Vol. 13; No. 1; 2022 

120 
 

◯ the lack of emotion which makes it dull for the listeners 

Question 3: What is the difference between pitch and intonation? 

◯ pitch indicates the change in vocal pitch, intonation indicates the 

highest/lowest point in a person’s voice 

◯ pitch indicates the high or low vocal point/degree, intonation indicates the 

change/changing process of the vocal pitch 

◯ pitch and intonation only differs based on perception 

Question 4: How many basic types of intonation are there? What are they? 

◯ 4 types: Flat tone / Rising tone / Falling tone / Falling-rising tone 

◯ 3 types: Rising tone / Falling tone / Falling-rising tone 

◯ 3 types: Rising tone / Falling tone / Upspeak tone 

Question 5: Default tones should be understood as: 

◯ the local dialect 

◯ the most used tone for each specific type of sentence  

◯ the difference in vocal tone between the male and the female  

Question 6: Choose the suitable default tones for the below types of sentences: 

                                       Falling intonation Rising intonation 

Wh-questions                             ◯                                  ◯ 

Yes-No questions                       ◯                                  ◯ 

Statements                                  ◯                                  ◯ 

Exclamations                              ◯                                  ◯ 

Question 7: In general, rising intonation is used to indicate… / falling intonation is 

used to indicate... (short answer) 

…………………………………… 

Question 8: To indicate uncertainty and doubts in the sentence: “I want to go to 

that party…, but…”, what type of intonation should be used? 

◯ Falling intonation 

◯ Rising intonation  

Question 9: To indicate enthusiasm and excitement in the sentence: “I'm glad to 

have you here”, how would you change your falling intonation? (short answer) 

……………………………………... 

Question 10: The variation in intonation used in tag questions indicates many 

different layers of meaning. Choose the suitable intonation for the meaning of the 

sentence: “You've been to London once, haven't you?” 



ACOJ- ISSN 1936-9859 AsiaCALL Online Journal  Vol. 13; No. 1; 2022 

121 

                                                                       Falling                   Rising 

The speaker genuinely wants 

to ask for information or is doubtful                 ◯                          ◯ 

about his/her knowledge. 

 

The speaker already has information                ◯                          ◯ 

and is certain about his/her knowledge. 

  Record the below sentences using the intonation that you have already known. 

1. Are you serious? 

2. Who was the girl you were talking to? 

3. I don't think it's safe to go there, is it? (the speaker is unsure) 

4. He has such a beautiful voice! 

5. I believe it is going to rain at the end of the week. 

6. Should I hand in my project? 

7. I'm going to the park. Would you like to come? 

8. He will be late again, isn't he? (the speaker already knows this) 

Add file here 

 

  

 

 

 

 

 

 


