INTRODUCTION
Aims of the study
This study aims to measure the effectiveness of applying computer software to teaching and learning English fricatives to English non-majored freshmen at HVU
The researcher conducted an experiment using computer software to teach challenging English sounds to Vietnamese students, aiming to assess the effects of this method on their learning outcomes.
In her experiment, she focused on English consonants due to their crucial role in word comprehension According to O’Connor (1986:24), mispronouncing a vowel may still allow the listener to identify the word, whereas a mispronounced consonant can lead to confusion with a different word.
The researcher focused on English fricatives for her experiment due to their difficulty for Vietnamese learners, particularly the sounds /θ/ and /ð/, which are absent in Vietnamese (Avery & Ehrlich, 1992: 155) These fricatives are also missing in many other languages, and even native English speakers can confuse /θ/ with /f/ in noisy environments Therefore, /f/ and /v/ were selected for comparison with /θ/ and /ð/, as such comparisons can provide valuable insights into the learning process.
Research questions
To achieve the above purposes, the study sets up the following research questions:
- What are the possible problems facing English non-majored freshmen at HVU in learning English fricatives?
- To what extent does computer software facilitate the teaching and learning of the four English fricatives /f/, /v/, /θ/ and /ð/?
- What are the students’ attitudes towards the technique?
Significance of the study
Currently, numerous free computer software options are available online, making the integration of pronunciation teaching with software less innovative The true value of this study lies in the careful selection of the software used and the effective methods of its application.
The researcher emphasizes that Pronunciation Power, focused on articulatory phonetics, and Praat software, designed for acoustic analysis, are essential tools for teachers to help students accurately identify English fricatives By utilizing Praat, students can audibly and visually analyze sounds through spectrograms and waveforms, comparing their own pronunciations with those of native speakers This approach enables students to recognize and self-correct their mistakes, making it an effective self-study method.
This self-study approach significantly enhances the teaching and learning of pronunciation in acoustic phonetics, particularly for learners who lack prior knowledge in the subject.
The research focused on English non-majored freshmen who lacked knowledge in acoustic phonetics, presenting an unexpected challenge for the researcher in determining her ability to address it.
Limitation of the study
This study concentrates on four English fricatives: /f/, /v/, /θ/, and /ð/, specifically in word-initial positions Additionally, the pronunciation is based on the accent of native speakers as represented in the Oxford Advanced Learner’s Dictionary, 8th edition, which serves as the standard reference.
Organization
The thesis consists of the following parts:
Chapter one: The introduction: states the study’s background, purposes, research questions, as well as the significance and design of the study
Chapter two : Literature review: discusses the definitions of related terms, and offers a review of the literature relevant to the present study
Chapter three : Methodology: focuses on the methodology employed in the study, including its participants and instruments, research problems, research design, and data collection procedures
Chapter four: Data analysis and discussion: discusses the results collected from the tests, questionnaires, and experimental teaching
Chapter five: Conclusion and suggestions: gives a brief summary of the thesis and some suggestions for better applications in the future.
LITERATURE REVIEW
English consonants
Studies of speech sounds show that it is not easy to define consonants exactly Overall, the most common view is to give a basic distinction between consonants and vowels
Consonants are sounds created by partially or completely obstructing airflow from the lungs through the vocal tract, as noted by Finegan (1994: 34) Additionally, Avery & Ehrlich (1992: 12) describe consonants as involving a narrowing in the mouth that leads to an obstruction of the airstream.
From the above definitions, we can make a conclusion that when we produce a consonant, there is a partial or complete obstruction of the air stream by a constriction of the speech organs
2.1.1The description of English consonants
According to Avery & Ehrlich (1992: 12), there are three factors to be considered:
- Place of articulation – where in the mouth the airstream is obstructed
- Manner of articulation – the way in which the airstream is obstructed
- Voicing – whether there is vibration of the vocal cords
2.1.1.1 English consonants in terms of place of articulation
According to Celce-Murcia, Brinton, and Goodwin (1996), the place of articulation refers to the specific position of speech organs used to produce consonant sounds There are six key places in the mouth where airflow is obstructed during consonant formation, as detailed by Avery and Ehrlich (1992) These places include bilabial, labiodental, interdental, alveolar, alveopalatal, and velar, corresponding to various consonant sounds such as /p/, /t/, /k/, /b/, /d/, /g/, and others.
Table 2.1: Classification of consonants according to place of articulation
2.1.1.2 English consonants in terms of manner of articulation
Understanding consonant articulation involves examining how speech organs interact, as highlighted by Celce-Murcia, Brinton, and Goodwin (1996) Avery and Ehrlich (1992) emphasize that the manner of articulation refers to the obstruction of the airstream during consonant pronunciation Various methods of airstream obstruction occur at different places of articulation within the mouth, which can be categorized into several types: stops (p, t, k, b, d, g), fricatives (f, θ, s, ʃ, v, ð, z, ʒ), affricates (ʧ, ʤ), nasals (m, n, ŋ), retroflex (r), lateral (l), and semi-vowels (w, y).
Table 2.2: Classification of consonants according to manner of articulation
2.1.1.3 English consonants in terms of voicing
Consonants can be classified based on their voicing, which refers to whether the vocal cords vibrate during the sound production Voiced sounds, such as b, d, g, and v, involve vocal cord vibration, while voiceless sounds, including p, t, k, and f, do not This classification is essential for understanding the phonetic characteristics of consonants, as detailed in the accompanying table.
Table 2.3: Classification of consonants according to voicing
Fricatives
Fricatives are consonants characterized by the escape of air through a narrow passage, resulting in a hissing sound (Roach, 1991) These continuant consonants can be produced without interruption as long as there is sufficient airflow (Roach, 1991) According to Fromkin & Rodman (1993), the narrowness of the oral passage creates friction or turbulence, which defines these sounds as fricatives.
2.2.1 The description of English fricatives
Table 2.4 illustrates eight fricatives characterized by their frictional qualities These sounds are produced when air from the lungs is forced through a narrow opening, resulting in varying types of friction The fricatives include bilabial [b], labiodental [v], interdental [θ, ð], alveolar [s, z], alveo-palatal [ʃ, ʒ], and velar [x].
In O’Connor’s description (1986: 25), the production of the consonants /f/ and /v/ involves raising the soft palate to prevent air from escaping through the nose, forcing it out through the mouth These sounds are classified as labiodental, where the bottom lip makes contact with the upper front teeth, creating a narrowing that causes slight friction as air passes through The key distinction between /f/ and /v/ lies in their strength: /f/ is a strong, voiceless, and longer consonant, while /v/ is a weak, voiced, and shorter consonant.
In the middle position, the /v/ is voiced, but short and weak The /v/ has short friction and that friction is caused by lip-teeth action whilst the /f/ has long friction
I feel fine / I feel vile four fans / four vans
In the final position, after a vowel, they have an effect on the length of the vowel The strong /f/ makes the vowel shorter; the weak /v/ makes the vowel longer
E.g: safe / save leaf / leave half / halve
The fricatives /θ/ and /ð/ are described with these features: (ibid, 29)
- The soft palate is raised so that all the breath is forced to go though the mouth
- The tip of the tongue is close to the upper front teeth: this is the narrowing where the friction is made
- The noise made by the friction for /θ/ and /ð/ is not very great, much less than for /s/ and /z/
- /θ/ is strong and /ð/ is weak
Between vowels, /ð/ is voiced, but short and weak; and /θ/ is always voiceless
At the end of words, /θ/ and /ð/ affect the preceding vowel in the same way as /f/ and /v/ (ibid, 31)
In the production of fricative sounds, the soft palate is elevated, directing airflow exclusively through the mouth The tongue's tip and blade approach the alveolar ridge, creating a significant narrowing away from both the teeth and hard palate The teeth remain closely positioned, resulting in increased friction, particularly for the /s/ sound, which is characterized by strong, prolonged voicelessness, in contrast to the weaker, shorter, and voiced /z/.
In initial position, /z/ is not a common sound
In middle position, between vowels, /z/ is voiced, gently and short /s/ is always voiceless
In final position, after a vowel, / s / makes the vowel rather shorter and / z / makes it longer (ibid, 32)
To pronounce the fricative pairs /ʃ/ and /ʒ/, the soft palate is raised, directing breath through the mouth while creating a narrow space between the tongue's tip and the alveolar ridge The front of the tongue is positioned higher compared to /s/ and /z/, and the lips are slightly rounded Notably, /ʃ/ produces a strong friction sound, whereas /ʒ/ results in a weaker one.
In initial position, /ʒ/ does not occur at the beginning of English words, but /ʃ/ quite frequently does
In middle position, /ʒ/ is voiced, /ʃ/ is always voiceless
In English, the sound /ʃ/ frequently appears at the end of words, while the sound /ʒ/ is uncommon, typically found only in a few French loanwords Notably, /ʒ/ lengthens the preceding vowel, in contrast to /ʃ/, which shortens it.
Articulatory Phonetics versus Acoustic Phonetics
Speech sounds can be objectively analyzed from three perspectives: the manner of production by human speech organs, the acoustic properties of sound waves, and their physical effects on the human ear This leads to a classification of phonetics into three main areas: articulatory, acoustic, and auditory phonetics.
Phonetics currently consists of two primary branches: articulatory phonetics and acoustic phonetics, as auditory phonetics has not yet been fully developed.
The traditional method of phonetic description focuses on 'articulation' by the speech organs Consonants are characterized by obstructed airflow, in contrast to vowels, which allow free airflow When the airflow obstruction is complete, the sounds produced are termed stops, whereas sounds with partial obstruction that create friction noise are known as fricatives.
Articulatory phonetics analyzes speech sounds based on key variables such as place of articulation, manner of articulation, voicing, and air passage According to the International Phonetic Alphabet (IPA), for instance, [p] is categorized as a bilabial, voiceless, oral stop, while [b] is a bilabial, voiced, oral stop Other examples include [f], a labio-dental, voiceless, oral fricative, and [m], a bilabial, voiced, nasal stop These characteristics serve as distinctive features that define the unique qualities of each sound.
Articulatory phonetics and acoustic phonetics differ significantly in their approach to understanding speech sounds Lyons (1968) noted that acoustic analysis reveals speech is not simply a sequence of discrete sounds, as the sound waves produced during speech vary in length and overlap Furthermore, the factors identified through acoustic analysis do not always align neatly with traditional criteria used by articulatory phoneticians to distinguish speech sounds While features such as voice, nasality, obstruction, and friction can be easily studied, identifying the place of articulation for sounds proves challenging From an acoustic perspective, distinctions between consonants at different places of articulation are primarily conveyed through contextual and transitional features in the sound wave, rather than inherent characteristics of the sounds themselves.
Acoustics as a branch of Physics
Acoustic phonetics, rooted in the principles of acoustics—a branch of physics—focuses on the scientific study of sound Key concepts in acoustics, established by 18th and 19th-century French mathematician Jean Baptiste Joseph Fourier, include wave motion, periodicity, frequency, and amplitude Waves, such as those in water, exhibit periodic movement, with wavelength representing the distance between recurring wave events While wavelength differs from the period, the periodicity of a waveform is determined by its frequency, measured in Hertz, which indicates how often an event repeats over a set time interval Additionally, amplitude reflects the wave's size, indicating the maximum deviation from a baseline, with larger amplitudes corresponding to "bigger" waves, measured in decibels Finally, the speed of a wave is referred to as its propagation velocity.
According to Fourier, the wavelength, frequency, and the propagation velocity are all related Algebraically the following relationship holds: λ f = c
Where λ = the wavelength, f = the frequency of the wave, and c = the propagation velocity
Application of acoustics in the study of speech sounds
Speech sounds are created by expelling air from the lungs through the mouth and occasionally the nose, with this airstream being essential for sound production (Fromkin, 1993:30) Similar to ocean waves, the characteristics of this airstream can be captured using a sound spectrograph, which produces spectrograms Acoustic phonetics utilizes these spectrograms to analyze various features of speech, including the place and manner of articulation, voicing, pitch, intensity, and loudness Notably, spectrograms can specifically reveal the place of articulation for different sounds.
The vocal tract can be viewed as a tube where sound waves travel from the larynx to the lips In this context, λ represents the distance from the stricture to the larynx, f denotes the frequency of sound waves, and c refers to the normal speech velocity The relationship among the place of articulation, the frequencies of the second formant on the spectrogram, and normal speech rate is described by Fourier’s formula: λ f = c (Lieberman & Blumstein, 1988: 24) When speaking at a typical rate, sounds produced closer to the larynx have a shorter λ compared to those produced farther away If the speech velocity (c) remains constant, an increase in λ results in a decrease in f, and vice versa.
Sounds produced at the bilabial position have a longer wavelength (λ) compared to those produced at the dental position Similarly, alveolar sounds exhibit a longer λ than alveo-palatal sounds Consequently, at a normal speech rate, labial sounds, with their longer λ, yield lower frequencies than dental sounds, which have shorter λ In contrast, alveo-palatal sounds, characterized by shorter λ, produce higher frequencies than alveolar sounds, which possess longer λ.
2.5.1.1.2 This calculation is proved correct in Ladefoged’s figure 8.9 (Ladefoged, 2011: 201) In this figure, the four fricatives / f, θ, s, ʃ/ are compared in the same linguistic context: [f, θ, s, ʃ] + [ai]
Ladefoged's analysis of the second formants of the sounds [f, θ, s, ʃ] in the words ‘fie’, ‘thigh’, ‘sigh’, and ‘shy’ reveals a clear frequency pattern The second formant of [f] in ‘fie’ is around 1000Hz, while [θ] in ‘thigh’ is approximately 1200Hz In ‘sigh’, [s] has a second formant near 1600Hz, and [ʃ] in ‘shy’ reaches about 1800Hz This indicates that as the sound moves farther from the larynx, the frequency of its second formant decreases.
2.5.1.2 Spectrograms can also provide information about the manner of articulation of a sound from the way the striations of that sound appear Fricatives appear as dark patches near the top of the spectrogram (Ladefoged, 2011:108) These dark patches consist of many striations occurring close together and without order, indicative of random noises Since these random noises occur in high frequency regions, fricatives have higher frequencies than other sounds aperiodic random noises of [ʃ]
5000 distinct dark bands of the vowel [i:] Figure 2.2: Spectrogram of the word ‘she’
The fricative [ʃ] appears in random noises while [i] appears as distinct dark bands [ʃ] occurs in high frequency regions while the vowel [i:] occurs in low frequency regions
To understand the articulation of fricatives compared to other consonants, we can examine the spectrogram of a stop consonant Stops are characterized by a complete closure of the vocal tract, which is visually represented in spectrograms as a white blank column indicating silence (Roach, 1991: 30) This distinct feature highlights the differences in sound production between stops and fricatives.
F re q u en cy ( H z) about the white blank column voice bar
Figure 2.3: Spectrogram of the word ‘about’
The stop[b] in ‘about’ appears as a white blank column before the formants of the diphthong [au]
2.5.1.3 Spectrograms can provide information about the voicing of a sound
Lieberman and Blumstein (1988) describe the acoustic characteristic of the voiced feature as the presence of low-frequency spectral energy or periodicity, resulting from vocal cord vibrations These vibrations manifest on a spectrogram as striations close to the baseline, commonly known as a voice bar (Ladefoged, 2011).
In figure 2.3, the term 'about' reveals numerous striations near the baseline, which indicate the presence of voice and are known as voice bars Additionally, spectrograms are valuable tools for analyzing sound, as they can provide insights into the pitch of the audio.
In voiced sounds, the information about pitch is provided by the computer: on the computer screen, it appears as a blue line above the voice bar near the base line
Figure 2.4: Spectrogram of the word ‘vine’
In the word 'vine,' the sound [v] is represented on the left side of the spectrogram, preceding the diphthong [ai], with its pitch indicated by the blue line positioned above the striations of the voice bar.
In voiceless sounds, the pitch is indicated by the upper formants of the sound, which occur in high frequency regions: the higher the frequencies, the higher the pitch.
Figure 2.5: Spectrogram of the word ‘fine’
The sound [f] in "fine" appears on the left side of the spectrogram, preceding the diphthong [ai] Its formants are positioned at higher frequencies compared to those of [v], and [f] also has a higher pitch than [v].
2.5.2 Besides spectrograms, computers also provide waveforms which analyze complex sound waves into relative amplitudes h ig h e r fr e q u e n ci e s lo w er f re q u en ci e s
2.5.2.1 Waveforms can provide information about the manner of articulation Let us compare the spectrograms in figures 2.2 and 2.3 with the waveforms of the same words
Figure 2.6: Waveform of the word ‘she’
In the waveform of ‘she’, [ʃ]appears as small variations before the larger variations of the vowel [i:]
Figure 2.7: Waveform of the word ‘about’
2.5.2.2 Waveforms can provide information about loudness According to Lieberman and Blumstein (1988: 27-28), loudness is our perceptual response to the amplitude of the speech signal When we listen to short sounds, the duration of the sound also influences our perceptual judgment of loudness However, amplitude i.e the air pressure of the speech signal, is the primary acoustic “correlate” of the percept of loudness
In figure 2.6, the word 'she' is made up of two distinct sounds: [ʃ] and [i:] The sound [ʃ] is a voiceless sound characterized by low amplitude and irregular air pressure variations, while [i:] is a voiced sound that is significantly louder, produced with vibrating vocal folds and regular air pressure pulses.
Thus, a sound with a large-amplituded waveform is a loud sound Let us consider the noun phrase ‘a spy’ for example
Figure 2.8: Waveform of the word ‘a spy’ (Ladefoged, 2011:158)
In the analysis of sound amplitudes, the vowel [ə] has a smaller amplitude compared to the diphthong [ai], making [ai] the loudest sound in the word "a spy." Additionally, the consonant [s] exhibits a soft sound due to its small amplitude, while the consonant [p] lacks any amplitude at the beginning and is inaudible during the closure Overall, the diphthong [ai] stands out as the most prominent sound in this context.
Computers enable the analysis of sound through spectrograms and waveforms Spectrograms break down speech into its component frequencies, offering clear visuals of articulation, voicing, and pitch In contrast, waveforms focus on the relative amplitudes of sound waves, providing insights into the manner of articulation and loudness.
Application of acoustics phonetics in the study of English fricatives
Research indicates that the key acoustic feature signaling the fricative manner of articulation, regardless of voicing or place of articulation, is the presence of aperiodic noise within the sound spectrum (Delattre, Liberman & Cooper, 1962).
2.6.1 Fricatives differ from each other in place of articulation There are four positions on the vocal tract where English fricatives can be produced:
- the labio-dental position where /f/ and /v/ are produced;
- the interdental or post-dental position where /θ/ and /ð/ are produced;
- the alveolar position where /s/ and /z/ are produced;
- the alveo-palatal position where /ʃ/ and /ʒ/ are produced
The differences between fricatives are influenced not only by their articulation points but also by the turbulence of air created during pronunciation When the tongue tip is raised against the alveolae or alveo-palatal region, it creates a partial obstruction that the rushing air stream encounters This air then meets an additional barrier formed by the lower row of teeth, resulting in turbulence as it breaks against their edges This intricate obstruction leads to the production of strident or sibilant fricatives such as /s, z, ʃ, ʒ/.
In the post-dental, interdental, and labio-dental positions, the airstream flows smoothly through the spaces between the upper teeth and the tongue tip or lip, while the lower teeth are concealed by the tongue Unlike sibilant sounds such as /s, z, ʃ, ʒ/, there is no additional barrier or turbulence of air The resulting sounds /f, v, θ, ð/ are characterized as mellow or non-sibilant.
For illustration, see the figure 2.1 of ‘fie, ‘thigh’, ‘sign’ and ‘shy’ quoted from Ladefoged’s figure 8.9
2.6.2 Fricatives also differ from each other in manner of articulation
Fricatives are produced by continuously and gradually releasing air through a constriction in the vocal tract The distinguishing factor between different fricatives is the duration for which a speaker maintains this airflow For instance, in Ladefoged’s figure 8.9, the duration of the [f] sound is approximately less than 100 milliseconds.
[θ] is about 100 ms, that of [s] is about 120 ms, and that of [ʃ] is about 150 ms Let us compare the fricative [f] in ‘fine’ with the fricative [θ] in ‘thigh’ for example
Figure 2.9: Waveform of [θ] in ‘thigh’
The sound [f] is shorter than [θ]
2.6.3 Fricatives differ from each other in pitch
All fricatives have high pitch As Ladefoged puts it, “fricative: random noise pattern, especially in high frequency regions” (Ladefoged, 2011: 204)
However, the strident fricatives are higher in pitch than the non-strident fricatives
Figure 2.11: Spectrogram of the word ‘fang’
Figure 2.12: Spectrogram of the word ‘thank’
The sound [f] is more strident than the sound [θ]
Figure 2.13: Spectrogram of the word ‘sank’
Figure 2.14: Spectrogram of the word ‘shank’
The sound [s] is more strident than the sound [ʃ]
Both [s] and [ʃ] are more strident than [f]and [θ]
The noise frequencies for the sounds [s] and [ʃ] are notably higher, ranging from 5000 to 6000Hz for [s] and 2500 to 3500Hz for [ʃ] In contrast, the sounds [f] and [θ] produce noise at lower frequencies, approximately between 2000 and 2500Hz Due to their greater acoustic intensity, [s] and [ʃ] generate darker patterns compared to [f] and [θ].
2.6.4 Fricatives differ from each other in loudness
Since amplitude is the primary acoustic correlate of the percept of loudness (Lieberman: 27-28) fricatives with large amplitudes are louder than those with smaller amplitudes
In examining the sounds [f] and [θ], it is evident that the amplitude of the sound waves for [f] in the word 'fine' is lower than that of [θ] in 'thigh' Consequently, this difference in amplitude results in [f] being perceived as softer than [θ].
2.6.5 Fricatives differ from each other in voicing
Voiced fricatives are paired with their voiceless counterparts, similar to stops and affricates However, unlike stops, the voicing in fricatives relates to loudness, as noted by Ladefoged He observes that sounds produced with vibrating vocal folds exhibit larger, regular air pressure pulses, while sounds without vocal vibrations display smaller and irregular air pressure variations (Ladefoged, 2001:7).
Let us consider the sentence “please pass me my book” for example
Figure 2.15: waveform of ‘please pass me my book’ (Ladefoged, 2011:215) The sound [z] in ‘please’ has larger amplitude than the sound [s] in ‘pass’
The corollary is all the voiceless fricatives such as /f, θ, s, ʃ/ are softer than and the voiced fricatives /v, ð, z, ʒ/
Computer software in teaching pronunciation
Computers have emerged as powerful tools in language teaching, significantly enhancing the learning experience Computer-assisted language learning (CALL) utilizes technology to support the presentation, storage, and assessment of educational materials, making it an effective method for language acquisition.
(http://www.qwertystudios.com/speech/tts-study/study-accurate- pronunciation/computer-assisted-language-learning.html)
Modern computer-assisted language learning (CALL) programs utilize a mix of text, images, sound, and video to facilitate language acquisition Users engage with the material through typing, mouse interactions, or voice commands Recent advancements in CALL software, including the integration of CD-ROMs, DVDs, and various technologies, enhance the teaching of pronunciation, grammar, vocabulary, and other essential language components.
According to Pellow (1995, cited in Pham Thi Thuy Van, 2006), the use of computer-generated descriptions of movies or videos can significantly enhance problem-solving skills The author suggests that computer aids and their application software, along with effective annotations and animations, enable teachers to present, practice, and produce language in diverse ways, ultimately enriching the language learning environment.
2.7.1 Introduction of some computer software to teach pronunciation
There is a number of software of teaching and learning pronunciation Among them, these five ones below are easily accessible and totally free to download on the Internet
2.7.1.1 EyeSpeak (http://www.eyespeakenglish.com/en/)
EyeSpeak, developed by Visual Pronunciation Software Ltd in New Zealand, utilizes native English speakers and everyday language to enhance learners' pronunciation, listening comprehension, and vocabulary skills in a natural way The software allows users to hear sound pronunciations, record their voices, and visually compare their speech with that of native speakers through mouth diagrams, waveform diagrams, and audio recordings However, a notable limitation is that while it provides pitch and timing information, it only indicates the pitch of words rather than specific sounds, and it lacks features related to the place and manner of articulation.
2.7.1.2 Sephonics (http://sephonics.findmysoft.com/)
Sephonics 1.0 is a Windows program that teaches learners the English phonetic alphabet, which is a subset of the International Phonetic Alphabet It includes seven different exercises for practicing English pronunciation and the phonetic alphabet, including a phonetic memory game to relax between the lessons There are also exercises where you learn to match a sound to a phonetic sign, transcribe from phonetic text to ordinary text This software is useful for those who want to focus on practicing phonetic alphabet
2.7.1.3 Ultimate phonics (http://spencerlearning.com/ultimate-phonics/free- program.html)
This program offers a structured approach to teaching pronunciation, speaking, and reading skills across various levels, from beginner to advanced It covers all phonics sounds and rules of English, making it an effective tool for enhancing learners' reading abilities.
2.7.1.4 Pronunciation Power (http://www.englishlearning.com/)
Pronunciation Power is designed to enhance American English pronunciation, making it suitable for both language labs and self-learners using PC or Macintosh multimedia computers It covers all 52 sounds of American English and features QuickTime videos demonstrating correct mouth positioning from both front and side angles The program is available in two levels, catering to different learning needs.
Pronunciation Power 1 is aimed at those at beginning to intermediate level, and comes with the 8-in-1 English Dictionary at no additional cost
Pronunciation Power 2 is for those at an intermediate to advanced level Recommended for all beginning to advanced students of English
This comprehensive program, crafted by ESL language experts, offers interactive drills and helpful exercises to enhance learning It includes movies that demonstrate correct tongue and lip positioning, making it an excellent resource for both ESL language labs and home use Highly recommended for effective language acquisition.
2.7.1.5 Praat (http://www.fon.hum.uva.nl/praat/)
Praat is a free acoustic analysis software developed by Paul Boersma and David Weenink from the University of Amsterdam It can read sounds recorded within the program or from other audio sources, generating wave graphs that display various characteristics such as articulation, pitch, loudness, and voicing Additionally, Praat allows users to isolate specific sound bites and filter frequencies through manual adjustments or scripting.
To meet the diverse needs of learners, appropriate computer software packages can be selected, such as Pronunciation Power and Praat in this study Pronunciation Power is ideal for English beginners, as it effectively illustrates the place of articulation, a crucial aspect of consonant pronunciation, through visual representations of articulator movements However, mastering the place of articulation alone is insufficient for accurate consonant pronunciation; therefore, Praat is utilized to offer comprehensive insights into manner of articulation, pitch, loudness, and voicing.
Previous study
The use of computer software for teaching pronunciation has gained attention globally, with various studies highlighting its benefits However, this innovative approach is relatively new in Vietnam, where research on teaching individual sounds to elementary learners, particularly through acoustic phonetics software, remains limited The following studies aim to connect with this research area.
A study conducted by Su Tseng Lee at Australian Catholic University examined the impact of Computer Assisted Language Learning (CALL) programs on Taiwanese students' English pronunciation The findings indicated that integrating CALL with traditional teaching methods could effectively tackle challenges faced by pronunciation instructors, such as low student motivation and proficiency Similarly, Ali Farhan AbuSeileek's research at King Saud University focused on the effectiveness of computer-based pronunciation instruction for EFL learners, revealing that it significantly enhanced their ability to perceive and produce correct stress patterns.
The English Resource Center of DELL has identified only two studies focused on teaching pronunciation through computer software One notable study, conducted by Le Thanh Tu in 2009, examined the effectiveness of using minimal pairs to teach speech sounds to first-year students at the University of Transport in Ho Chi Minh City This research utilized four different computer software packages.
Pronunciation Power 1, Pronunciation Power 2, Lose Your Accent in 28 Days and
A study involving 54 students from class CN07B utilized Praat for data collection, revealing findings that are highly relevant to the student population Additionally, a second study conducted by Le Tan Phuoc in 2011 further explored the application of this software.
This study utilizes Praat to analyze and compare challenging consonants in Vietnamese and English, highlighting their differences and similarities The researcher applied these insights to enhance the teaching of consonants, aiming to improve the speaking skills of Vietnamese learners The findings emphasize the importance of raising awareness about these significant phonetic differences to help learners overcome the tendency to substitute Vietnamese phonemes for their English counterparts.
This study uniquely focuses on utilizing computer software to teach English fricatives, enabling students to learn proper articulation while also identifying and correcting their own mistakes.
Conceptual framework
According to Avery & Ehrlich and Ladefoged, five key features are essential for assessing English fricatives: place of articulation, manner of articulation, pitch, loudness, and voicing Utilizing these criteria, the researcher conducts experimental teaching to evaluate and grade students' pronunciation.
Summary
This chapter explores articulatory phonetics, detailing the characteristics of English consonants, including their place and manner of articulation, as well as voicing It provides an in-depth examination of English fricatives and contrasts articulatory with acoustic phonetics, highlighting the role of acoustic phonetics in analyzing speech sounds and specifically English fricatives The chapter also reviews the integration of computer software in pronunciation teaching, introducing various tools designed for this purpose Additionally, it surveys related studies that demonstrate the effectiveness of computer software in enhancing pronunciation instruction The insights gained from these concepts have inspired the researcher to establish the objectives of her study and identify potential solutions to relevant challenges.
METHODOLOGY
Setting
The research conducted at HVU involved 92 English non-majored freshmen from two information technology classes, 11CT and 11CDCT The selection of these students was convenient for the researcher, who is affiliated with the university, allowing for easier permission from the School Administering Board At the start of the 2011-2012 school year, students underwent a placement test administered by the English Department, which classified them into beginning level classes based on their results A pre-test conducted before the experimental teaching on October 24, 2011, confirmed that both groups had similar English proficiency levels The 47 students in class 11CT and 45 students in class 11CDCT were divided into an experimental group, which received pronunciation instruction using computer software, and a control group, which was taught without the aid of technology.
The Foreign Language Department of HVU has selected "Interchange" – Book 1 (third Edition) by Richards J C (2005) as the required course book for all freshmen This textbook is designed to enhance integrated skills, including listening, speaking, reading, and writing The English curriculum for non-English major first-year students comprises seventy-five periods, with each unit structured into six forty-five-minute sessions, allocating only twenty minutes specifically for pronunciation practice.
The study focused on students from classes 11CT and 11CDCT, who were considered to have comparable English proficiency These students faced similar challenges in recognizing and producing English fricatives, akin to other non-majored freshmen at HVU.
The tables below show the population of the samples in detail
Age Frequency Percent Cumulative Percent
The study focused on first-year students, with a significant majority (68.1%) being eighteen years old, indicating that most participants had recently completed high school and retained their knowledge from English lessons.
In an information technology class, the majority of students are male, comprising 85.1% compared to 14.9% female students This gender disparity influences the speed at which individuals access and utilize computer software, with males typically gaining proficiency more quickly than females Consequently, this trend results in a higher percentage of male students in technology-related fields.
Table 3.3 indicates that a significant majority of students, accounting for 63.8%, had studied English for seven years during high school, while only 6.4% had more than ten years of experience This suggests that the students possessed a sufficient foundational understanding of English to participate in the experimental teaching.
Table 3.3: Years of learning English
Research problems
Despite all students majoring in information technology, many lacked personal computers, an unexpected shortcoming that hindered their learning and negatively impacted their attitudes towards education.
The students were unfamiliar with acoustic phonetics and required an introduction to essential concepts, including spectrograms, waveforms, frequency, and amplitude They engaged in practical exercises to understand the second formant, as well as the aspects of articulation, pitch, loudness, and voicing.
Research design
The research included the following steps which made up the experiment
First step: The researcher began her experiment by teaching a class in which she taught her students how to pronounce /f, v, θ, ð/ with the use of two computer software
Following the experimental teaching phase, tests were administered to evaluate the students' achievements These tests included recordings of the students' pronunciation of various sounds, which served as the primary data for the experiment.
In the third step, the researcher gathered and processed all data using computer software The fourth step involved comparing the spectrograms and waveforms of sounds recorded from students with those of native speakers, as found in the Oxford Advanced Learner’s Dictionary CD-Rom 8th edition This comparison was based on five criteria: place of articulation, manner of articulation, pitch, loudness, and voicing, which reflect the distinctive features of English sounds as noted by Ladefoged in "A Course in Phonetics." Ladefoged emphasized that sounds of the same length can vary in pitch, loudness, and quality, with 'quality' being interpreted through the fundamental characteristics used by phoneticians to describe English consonants.
Fifth step: The researcher quantified the degrees of correctness in the students’ performances and decided on a scale of grading
Sixth step: The results of the grading were analyzed qualitatively before they were processed with the aid of SPSS.
Research instruments
The instruments used in this research included:
The researcher believed it was necessary to carry out two questionnaires (see appendix
Questionnaire 1 was designed to identify the challenges faced by English non-majored freshmen at HVU in learning English fricatives, while Questionnaire 2 sought to explore the students' attitudes towards the teaching techniques employed.
Nunan (1992) emphasizes that questionnaires are effective tools for gathering data from large groups, while Seliger and Shohamy (1989) highlight their utility in collecting information on less observable phenomena, including attitudes and self-concepts, as well as the processes involved in language use.
To ensure clarity and reliability in responses, the questionnaires were composed in Vietnamese The initial questionnaire, administered on October 10, 2011, aimed to gather information about the students' backgrounds and identify their challenges with English fricatives.
The article outlines a study consisting of 14 questions categorized into three sections The first section gathers students' personal information, while the second part focuses on eight questions regarding their past experiences with learning English pronunciation, including challenges with English sounds and fricatives Finally, the last section explores the students' potential for mastering English fricatives through experimental teaching methods.
In December 2011, a second questionnaire comprising 10 questions was administered to students to assess their evaluations and attitudes toward an experiment The initial three questions focused on students' perceptions of learning English fricatives through computer software The following four questions aimed to gauge their self-assessment of improvement, while the remaining questions sought their opinions and suggestions for enhancing English pronunciation learning.
The experiment was carried out for thirteen weeks from October 3 rd to December 26 th ,
2011 It included the pre-test and the post-test which will be described in detail as follows
A pre-test was conducted to assess the recognition and pronunciation abilities of English fricatives in both the control and experimental groups Prior to the experimental teaching, both groups took the same pre-test, designed and scored by the researcher (refer to appendix 1).
The pre-test consisted of two components: listening and pronunciation This dual approach aimed to assess the students' ability to recognize specific sounds as well as their capability to produce those sounds accurately.
The listening section included three exercises designed to assess students' phonetic recognition skills The first exercise focused on identifying words that start with the sounds /f/, /v/, and /p/ The second exercise compared words beginning with /θ/ and /ð/ to those starting with /t/, /d/, and /f/, /v/ Finally, Exercise 3 featured 20 sentences containing words that began with /f/, /v/, /θ/, and /ð/ All audio materials were recorded by native speakers prior to the test.
The pronunciation assessment aimed to evaluate how students articulated four fricative sounds at both the word and sentence levels prior to the experimental teaching Their pronunciations were subsequently recorded and assessed for grading.
The total score of the test was 10, 5 points for listening and 5 points for pronunciation
In the listening test, the students got 0.16 point for each correct answer in Part I, II, and III In the pronunciation test, the students got 1 point for each feature
The post-test assessed students' understanding of the material covered during the experimental teaching, focusing on the effectiveness of computer software in enhancing the learning of the four English fricatives It comprised two sections: a recognition test and a pronunciation test.
The test was designed to mirror the pre-test format, assessing students' progress in recognizing and producing the sounds /f, v, θ, ð/ by the experiment's conclusion An American native speaker pronounced all words and sentences live to evaluate students' recognition abilities.
The researcher played the role of the teacher and observer during the experimental teaching, and also the scorer of the students’ achievements through the tests
In the pre-test and post-test listening assessments, an American native speaker and English teacher provided live readings of words and sentences starting with the sounds /f, v, θ, ð/ This collaboration aimed to evaluate the students' ability to recognize these specific phonetic sounds.
The researcher used two software packages as a tool to facilitate the teaching of the four English fricatives in the experimental group
The first was “Pronunciation Power I & II”, which was developed by the English
Computerized Learning Inc offers innovative English learning solutions through their website, [englishlearning.com](http://www.englishlearning.com/) One of their notable products is "Praat," developed by Paul Boersma from the University of Amsterdam, which can be downloaded at [fon.hum.uva.nl/praat](http://www.fon.hum.uva.nl/praat/).
The software programs 'Pronunciation Power 1' and 'Pronunciation Power 2' were utilized to enhance students' awareness and improve their perception of specific sounds These interactive tools featured moving visuals that demonstrated the articulatory movements involved in producing the fricative sounds /f, v, θ, ð/.
The second software ‘Praat’ was applied to help the students make their recordings and analyze the recorded sounds
The software enabled students to access both audio and visual representations of their voices, enhancing their learning experience Additionally, it provided samples of standard pronunciation from native speakers through The Oxford Advanced Learner’s Dictionary CD-Rom, further aiding in their language acquisition.
8 th edition The students could compare the spectrograms and waveforms of their voices with those of native speakers to target specific problem areas in their own pronunciation
First step (October 3 rd 2011): The teacher introduced the experimentation and its purpose The objective of this step was to get the students’ cooperation
Data collection procedure
The study utilized two questionnaires to gather data from students, with Questionnaire 1 administered at the beginning of the experiment to 92 participants across both groups At the conclusion of the experiment, Questionnaire 2 was conducted with 47 students from the experimental group, allowing the researcher to collect and analyze the responses from both sets of questionnaires.
The study assessed the pronunciation improvement of experimental students through pre- and post-tests conducted on both experimental and control groups at the beginning and end of the teaching period The researcher analyzed the test results and compiled statistical data to highlight the differences in listening and pronunciation abilities between the two groups.
Data from spectrograms and waveforms were gathered during pronunciation tests conducted at the start and conclusion of the experiment Students were instructed to maintain the microphone close to their mouths while reading words and sentences from the test sheet Their performances were recorded and evaluated based on five specific criteria.
Treatment of data
The responses from students to both questionnaires were initially gathered and subsequently analyzed using SPSS software The results, including mean values, percentages, and frequency counts, were organized and displayed in tables and charts for clarity.
The results of the two tests were analyzed quantitatively The scores of the two tests were calculated and synthesized through the Statistical Package for the Social Sciences (SPSS)
A t-test for independent samples was conducted to assess the differences in mean scores between two groups, evaluating the impact of a specific treatment on their performance in two tests.
In this study, the test was utilized to identify significant differences in mean scores between the experimental and control groups during both the pre-test and post-test phases The null hypotheses being examined include: (1) the absence of significant differences in variances between the experimental and control groups in the pre-test, and (2) the absence of significant differences in variances in the post-test.
The study proposed two two-tailed experimental hypotheses: first, that a significant difference exists in the variances of the experimental and control groups during the pre-test; and second, that a significant difference is also present in the variances of these groups in the post-test.
In this study, a standard significance level of 5% (alpha level of 0.05) was employed for all statistical tests, as commonly used in educational and social science research (Allison, 2002: 120) If the significance values exceed 0.05, the null hypothesis is accepted, indicating no significant difference in pronunciation ability between the two groups Conversely, if the significance values are 0.05 or lower, the null hypothesis is rejected, suggesting a significant difference in pronunciation competence between the groups.
The analysis of students' pronunciation enhanced the reliability and validity of the research To evaluate their performances accurately, the researcher quantified the levels of correctness prior to assigning grades.
3.6.3.1 Quantification and grading of place of articulation
The second formant is crucial for determining the place of articulation in speech sounds To assess a student's articulation accuracy, we compare their second formant frequency to that of a native speaker A difference of only two digits is deemed insignificant; for instance, if a student's [θ] in [θai] is measured at 1180 Hz or 1230 Hz, while a native speaker's [θ] is at 1200 Hz, the student's pronunciation is considered correct, earning them a point However, if the student's [θ] falls around 1150 Hz, further evaluation is necessary.
Hz and 1250 Hz, he would get 0.75 point If it occurred in the neighborhood of 1100
Hz and 1300 Hz, he would get 0.50 point If it occurred in the neighborhood of 1050
Hz and 1350 Hz, he would get 0.25 point If it occurred in the neighborhood of 1000
Hz and 1400 Hz, he would get zero point
At 1150 Hz, the student’s [θ] was a little forward; at 1100 Hz, it was forward; at 1050
Hz, it was quite forward; at 1000 Hz, it was too forward and sounded more like [f] At
1250 Hz, the student’s [θ] was a little retracted; at 1300 Hz, it was retracted; at 1350
At 1100 Hz, the [θ] sound is articulated more forward in the vocal tract compared to the standard [θ], while at 1300 Hz, it becomes more retracted This phenomenon can be explained using Fourier’s formula: λ f = c, where the wavelength (λ) varies inversely with frequency (f) when the speed of sound (c) remains constant Therefore, a [θ] sound at a higher frequency, such as 1300 Hz, is produced at a shorter wavelength, resulting in a more retracted position In contrast, a [θ] sound at a lower frequency, like 1100 Hz, is pronounced at a longer wavelength, leading to a more advanced articulation in the vocal tract relative to the standard [θ].
3.6.3.2 Quantification and grading of manner of articulation
Fricatives are produced by forcing air through narrow gaps, and assessing a student's articulation involves evaluating how they initiate, maintain, extend, and conclude the fricative sounds.
To achieve a smooth onset in fricative sounds, the student should gradually and continuously rush the air, earning 0.25 points for this initial smoothness Continuing to maintain a steady airflow will result in a continuous and regular sound course, allowing the student to earn an additional 0.25 points Prolonging the fricative will be reflected in the waveform's horizontal line, indicating the duration of the sound, as noted by Lieberman and others.
According to Blumstein (1988), the duration of a stop in natural speech typically ranges from 25 to 55 milliseconds, while fricatives last about 100 milliseconds A student can earn an additional 0.25 points if their fricative duration falls between 80 and 100 milliseconds Furthermore, the student can receive a final 0.25 points if the waveform of their fricative exhibits a smooth decay with a decrease in amplitude.
3.6.3.3 Quantification and grading of pitch
In analyzing voiced and voiceless sounds on a spectrogram, pitch is represented by a blue line near the baseline for voiced sounds and by major frequency peaks for voiceless sounds Grading pitch involves comparing the placement of the blue line and frequency peaks on the student's spectrogram to those of a native speaker If they align in the same frequency region, the student earns one point However, if discrepancies exist, similar to the grading of place of articulation, minor differences within two digits are deemed insignificant, while larger differences result in scores of 0.75, 0.50, 0.25, or zero, depending on the extent of the deviation.
3.6.3.4 Quantification and grading of loudness
Loudness, our perceptual response to speech signal amplitude, is crucial in grading speech clarity (Lieberman and Blumstein, 1988:27) Fricatives such as /f/, /v/, /θ/, and /ð/ typically have an amplitude about one-fifth that of vowels If a student’s waveform amplitude matches that of a native speaker, they receive one point However, if their amplitude is twice that of the native speaker, they earn 0.75 points; three times larger results in 0.5 points, and four times larger yields 0.25 points Should the amplitude exceed four times that of the native speaker, often matching the amplitude of the subsequent vowel, the student receives zero points.
3.6.3.5 Quantification and grading of voicing
The grading of voicing in fricatives is determined by the appearance of the voice bar on the spectrogram A thick, continuous voice bar indicates a wholly voiced sound, earning the student one point If the voice bar appears in shreds but covers the entire duration, the sound is partially voiced, resulting in 0.75 points A shredded voice bar that extends over only part of the fricative is deemed not voiced enough, granting 0.5 points When the voice bar is fragmented and covers only a fraction of the fricative, it is considered poorly voiced, resulting in 0.25 points Finally, if no voice bar is present during the pronunciation of a voiced fricative, the student receives zero points.
When a voiceless fricative is articulated, the spectrogram should exhibit no voice bar near the baseline A student demonstrating a clear absence of a voiced bar would earn one point, while the presence of faint dark striations above the baseline would result in a score ranging from 0.75 to zero points, depending on the darkness and quantity of the striations.
Summary
This chapter outlines the methodology used to teach 92 English non-majored freshmen how to pronounce the four English fricatives [f, v, θ, ð] through computer software The students were selected based on their similar English proficiency and common challenges in recognizing and producing these sounds The study involved preparatory steps followed by experimental teaching, with outcomes assessed through pre-tests, post-tests, and questionnaires Data collected from these performances will be analyzed in the subsequent chapter.
ANALYSIS AND DISCUSSION
Results
Questionnaire 1 was delivered to 92 students in the experimental and control group at the beginning of the experimental teaching Of the 14 questions in the questionnaire, the first three have already been described in Chapter 3 and the rest of 11 questions were subdivided into two main parts to serve the research purpose The following sub- sections will present in detail the analysis of the data of the results
4.1.1.1 Students’ opinion about the teaching and learning of English pronunciation
Question 4 to 11 is to find out about the pronunciation teaching and learning programs at the students’ high school This section helps the researcher decide the appropriate techniques to teach the fricatives and how to apply computer software to improve students’ pronunciation of the four fricatives
Question 4 investigates the students’ period of learning English per week The students’ responses ranged from 2 to 8 periods Among these, more than one third of the students (38%) studied English 5 periods per week
Last year, how many English periods did you have a week? Frequency Percent
Table 4.1: Students’ periods of learning English
Question 5 helps to confirm whether the students’ previous teachers of English really taught pronunciation to them in class Although there were at least two periods of
English per week, only 44.7% of these were used to teach pronunciation, whereas most of them (55.3%) used to teach pronunciation
A survey revealed that 95.5% of high school students in the English program received grammar instruction, while 31.8% focused on vocabulary Additionally, 22.7% of students were taught writing skills, and 18.2% received instruction in reading skills, indicating a lack of emphasis on pronunciation in their curriculum.
Grammar Vocabulary Writing skill Reading skill Listening skill Count
Table 4.2: The concentration of English program at their schools
Question 7 was designed to investigate the methods of teaching pronunciation in class
According to Chart 4.2, a significant majority of respondents (75.9%) preferred to ask teachers to repeat information after listening, while 20.7% opted for repeating after listening to CDs or cassettes In contrast, a minimal percentage (3.4%) utilized computer software as a learning method.
Ask you to repeat after listening to teachers, 75.9%
Ask you to repeat after listening to CDs/ cassettes, 20.7%
Use computer softwares to teach, 3.4%
Chart 4.2: Methods of teaching English pronunciation in class
Question 8 collects information about the frequency of pronunciation teaching by their previous teachers Nearly two thirds answered that they were sometimes taught pronunciation, 33.3% was often and only 7.8% of the respondents was very often
Chart 4.3: Frequency of pronunciation teaching in class
Question 9 was to find out if students used to learn English by using computer software The answers show that nearly all the students (91.3%) did not have a chance to learn English with the aids of computer software
Have you ever learned English by using computer software? Frequency Percent
Table 4.3: Students’ chance to learn English by using computer software
Question 10 aimed at investigating students’ knowledge of basic concepts about phonetics Most of them did not know the notions of fricatives (97.7%), stops (95.5%), voiced (77.8%), voiceless (75.6%), spectrogram (69.6%), place of articulation (67.4%), waveform (63.8%), manner of articulation (57.8%), frequency (46.8%) and loudness (36.2%) Students were non English major; therefore most of them did not know the basic concepts of phonetics
Fr ic at iv es
S pe ct ro gr am
M an ne r o f a rti cu la tio n
P la ce o f a rti cu la tio n
Chart 4.4: Students’ perception of some basic concepts about phonetics
In response to question 11 regarding the sources of knowledge for the concepts mentioned in question 10, over 62.9% of students indicated they learned these concepts in Physics, while nearly 31.4% attributed their understanding to in-school English programs, and a small percentage reported learning from other English programs.
How do you know about the concepts mentioned in question 10? Frequency Percent
Learned in the in-school
Learned in other English programs 2 2.2 5.7 100.0
Table 4.4: The sources students learned the concepts
4.1.1.2 Students’ English studying habits at home
This section was devised to find out how the students practiced pronouncing English sounds outside their classes
Question 12 finds out how much time per day the students spent on self-studying English Nearly haft of students (56.5%) spent half an hour per day, more than one third used one hour and 8.6% spent two hours or more learning English
Chart 4.5: Students’ time spent on learning English per day
Question 13 was intended to survey students’ frequency of practicing pronunciation while studying English As observed from the chart below, most of the students did not often practice pronunciation: 31.9% of them never practiced, and only 63.8% sometimes practiced
Chart 4.6: Students’ frequency of practicing pronunciation
To explain the reason for not practicing frequently, the following table shows their problems of studying pronunciation
What are your problems in studying English pronunciation? Frequency
I did not know how to pronounce a sound correctly 22
I could not remember how to pronounce a sound 19
Pronunciation was so difficult to learn 16
I did not feel confident in pronunciation 13
I did not know how to pronounce difficult sounds like /θ/ and /ð/ 8
My teachers did not focus much on pronunciation 5
My teachers did not show how to correct our mistakes in pronunciation 3
Table 4.5: Students’ problems in pronunciation
Before the experimental treatment, students in the experimental group shared their feelings about previous English teaching programs and their study habits Their responses revealed that a significant challenge was their inability to pronounce sounds correctly Therefore, it is essential to assist students in mastering proper pronunciation, as will be demonstrated in the following section.
Questionnaire 2 was delivered to the 47 students in the experimental group in regular class time after the experimental teaching This section aims to analyze and discuss the data to investigate the students’ attitudes towards the application of computer software to teach the four fricatives as well as the learning of pronunciation
Question 1 was to find out the students’ interests in learning the four fricatives with computer software The results in the table show that 27.7% of the students contended that they were very interested while 63.8% of them were interested There were only
8.5% responses with not really interested These figures reveal that their preference for the learning pronunciation with computer software was great
Ss’ interests in learning the pronunciation of the four fricatives with computer software after ET Frequency Percent
Table 4.6: Students’ interests in the four fricative learning
In response to the reasons for their interest in learning with computer software, 38 participants noted that this approach enabled them to identify and correct their mistakes Additionally, two respondents highlighted improvements in their pronunciation, while four others appreciated the novelty of the method, stating that it made learning periods more engaging and effective.
Can you give the reason(s) for question 1 Frequency
This method made the class more interesting and more effective 2
I did not have access to a computer for private practice 1
This method helped me recognize my mistakes and correct them 38
Table 4.7: Ss’ reasons for their interests of learning with computer software
Discussion
In this section, the analysis of the data collected through the questionnaires presented in Section 4.1 is discussed in relation to the aim of the study
In the first questionnaire, responses indicated that English classes primarily focused on teaching reading, writing, listening, and oral fluency, while pronunciation received less attention According to Celce-Murcia, Brinton, and Goodwin (1996), grammar and vocabulary are more commonly understood by language teachers than pronunciation, leading to its neglect in instruction Students reported difficulties in pronouncing specific sounds, such as /θ/ and /ð/, highlighting a need for targeted instruction in these challenging areas Traditional methods based on articulatory phonetics fail to meet learners' needs, especially for those studying information technology, indicating a demand for more engaging and relevant teaching approaches.
In the second questionnaire, students expressed a strong preference for using computer software to enhance their pronunciation of the fricatives /f/, /v/, /θ/, and /ð/ Most participants reported significant improvements in their pronunciation and expressed a desire to practice additional sounds with the software in the future These findings align with Lieberman and Blumstein’s (1988) assertion that the study intersects with various disciplines, including physics, mathematics, and computer science, enabling students to grasp concepts related to sound physics and speech production Furthermore, Brett (1996) highlights that learners show notably positive attitudes toward multimedia learning, particularly among those who may struggle more with traditional methods.
4.2.2 Discussion on the pronunciation tests
Students often mispronounced the sounds /f/ and /v/ due to incorrect articulation Specifically, they tended to add a slight /p/ sound when pronouncing /f/, as they inadvertently shifted their upper lip forward after placing it against their lower lip This movement interrupted the airflow, resulting in a [p] sound following [f] The issue arose from transitioning from a labio-dental to a bilabial position However, this error was easily corrected by instructing students to maintain their initial labio-dental position while articulating /f/.
The mistakes in pronouncing /θ/ and /ð/ were predominantly those made in manner of articulation, though mistakes in place of articulation were also involved
Usually, at first, the students would place their tongue tip against the alveolae – a wrong position! Then they released a strong airstream from this position: the outburst was heard as /t/ or /d/
Students learned to produce the sounds /θ/ and /ð/ by directing a continuous airstream over the tongue positioned between the teeth Initially, they succeeded in creating these fricative sounds, but after recognizing their achievement, they became complacent and reverted to their previous habit of resting the tongue tip against the alveolar ridge before transitioning to the stop sounds /t/ and /d/.
Figure 4.23: Waveform of a student’s [θ] in ‘thick’
The fricative [θ] at the beginning was followed by the stop [t h ]
The student's mistake indicates a solid understanding of how to produce the English fricative [θ] However, after completing the task, his tendency to pronounce [θ] as a heavily aspirated plosive [t h ] disrupted his performance, leading him to articulate the fricative [θ] followed by the plosive [t h ].
To effectively teach Vietnamese students the pronunciation of /θ/ and /ð/, teachers should go beyond basic instructions like "rushing your airstream over the surface of the tongue." It's essential to guide students on how to transition smoothly from these sounds to the subsequent vowel sounds.
In the above case, a student made a fricative before a stop In the following case, another student made a stop before a fricative – a different strategy! the stop [t h ] the fricative[θ ]
Figure 4.24: Waveform of another student’s [θ] in ‘thick’
The spike at the beginning marks the release of a stop and the fricative follows the stop
In the illustration, the student unintentionally produced a stop sound while preparing to create a prolonged airstream He initially rested his tongue tip against the alveolae to gather strength When he attempted to start, he opened his mouth to reposition his tongue, but the built-up air pressure behind it was released prematurely, resulting in an unintentional plosive sound before the intended fricative.
The unintentional stop resulted in a minor spike, with a brief silence preceding the outburst As illustrated in figure 4.24, the student's waveform demonstrates his ability to produce a clear fricative [θ] following the stop, as seen in the burst of [t h] in the word 'thick.'
Figure 4.25: Waveform of a student’s [θ] in ‘thick’
To help the students rectify this mistake, the teacher could simply tell them to take it easy and should not make their preparation too long
Students experienced challenges with voiceless sounds, while voiced sounds posed less difficulty Research by Lieberman and Blumstein (1988) indicates that voiced sounds require significantly less airflow compared to voiceless sounds, with sounds such as [h] and [s] necessitating the highest airflow Additionally, observations of the glottis reveal that the larynx opens wider for producing [s] than for [p], suggesting that speakers intentionally adjust their glottis to enhance airflow for voiceless sounds (Sawashima & Miyazaki, 1973; Lieberman and Blumstein, 1988).
Linguistic studies indicate that producing voiceless sounds requires native speakers to consciously exert effort to open their vocal cords and push more air through the vocal tract Therefore, if test results reveal that students struggle with voiceless sounds while managing voiced ones, it likely stems from insufficient effort in articulating those voiceless sounds.
To evaluate the accuracy of the argument, we should consider additional sound characteristics If students struggle with producing voiceless sounds due to insufficient airflow, they are likely to encounter difficulties with other sound features that require similar effort.
Pitch is a key characteristic of sound that requires effort to produce The rate of vibration of the vocal cords, which is influenced by the level of tension, determines the pitch of the sound (Lyons, 1968:103) Therefore, the production of pitch is directly related to the amount of tension applied.
According to Jakobson, tense phonemes are characterized by longer sound intervals and greater energy compared to lax phonemes This implies that the degree of tension in phonemes is linked to the amount of energy or effort exerted Lyons and Jakobson suggest that producing high-pitched sounds necessitates a higher degree of tension and effort Therefore, if students struggle to produce voiceless sounds due to insufficient effort, this lack of exertion may also result in lower-pitched sounds.
The relationship between effort and sound loudness is significant, as highlighted in "A Course in Phonetics" by Ladefoged, who connects loudness to intensity, defined by the Oxford Advanced Learners’ Dictionary as the strength of something This suggests that sounds produced with greater intensity are articulated with more strength, implying that the effort required for voicelessness and high pitch also contributes to sound loudness Consequently, a correlation likely exists among voicelessness, pitch, and loudness, indicating that higher levels of voicelessness and pitch correspond with increased loudness, and vice versa.
However, the students’ performances of [f] in the pre-test proved the opposite: of the
47 performances, there were 39 students pronouncing their [f] louder than the native speaker versus 08 students pronouncing their [f] softer than the native speaker
The statistics in 4.1.5.1.5 show that the strength or the effort with which one pronounces a sound does not equate the loudness of that sound That is, the meaning
‘strength’ given by the Oxford Dictionary is only the meaning of that word in general English Phonetically, a loud sound is not a strong sound In his “Preliminaries to
In "Speech Analysis," Jakobson highlights the significance of duration in differentiating between tense and lax sounds, suggesting a connection between this feature and the prosodic contrast of long and short durations Thus, duration plays a crucial role in understanding the distinction between these two sound categories.
Summary
This chapter presents the findings from two questionnaires, as well as pre- and post-test results, revealing that students exhibited positive attitudes towards the method Analysis of native speakers’ and students’ spectrograms and waveforms identified common pronunciation mistakes made by students The pre-test and post-test results indicated that the experimental group showed greater improvement in pronunciation compared to the control group Additionally, the discussion highlighted the factors contributing to these mistakes and proposed strategies for rectification.