Pronunciation Accuracy Research Articles

Using Speech Recognition Algorithms to Improve Listening Training in College English Listening Instruction

English listening proficiency plays a crucial role in language learning and communication. Speech recognition algorithms can significantly enhance listening training in college English instruction by providing personalized and interactive learning experiences. These algorithms can transcribe spoken English passages, allowing students to receive immediate feedback on their comprehension and pronunciation. Additionally, they can generate exercises tailored to individual proficiency levels, offering targeted practice in areas needing improvement. Incorporating speech recognition technology into listening instruction not only promotes active engagement but also enables instructors to track students' progress more effectively. This paper proposed a novel approach leveraging Generative Adversarial Networks (GANs) within an Optimized Edge Computing (OEC) framework to enhance College English Listening Instruction through speech recognition. Traditional methods for English listening instruction often face challenges in providing authentic and personalized learning experiences. To address these limitations, we harness the power of GANs to generate synthetic speech data that closely resembles real-world English speech patterns. By integrating GANs with OEC, we achieve efficient processing of speech data at the edge, minimizing latency and bandwidth consumption while ensuring real-time feedback and interaction. Through extensive experimentation and analysis, demonstrate the superiority of our GAN-OEC framework over traditional baseline models and other speech recognition algorithms. Results demonstrated that the proposed model achieves an average accuracy of 90.6%, a fluency score of 8.5 out of 10, and a pronunciation accuracy of 84.6%. These results highlight the transformative potential of GAN-OEC in revolutionizing English instruction and language learning outcomes in educational settings.

EXPLORING THE USE OF ARTIFICIAL INTELLIGENCE IN PROMOTING ENGLISH LANGUAGE PRONUNCIATION SKILLS

This research examines the potential of artificial intelligence (AI) in enhancing learners' English pronunciation skills. The participants were 78 English language learners at the elementary and pre-intermediate levels, and 19 experienced English language teachers. The researchers employed a mixed-method approach, integrating both quantitative and qualitative data collection techniques, to comprehensively assess the efficacy of AI-based pronunciation aids and gauge learners' perceptions. During a course of two months, the experimental group used Listnr and Murf AI tools, while the control group adhered to a conventional instruction. The data were taken from pre- and post-test pronunciation scores, questionnaire responses, and interviews with the individuals. The findings of the pre- and post-test indicated that the participants in the experimental group had significant improvements in their pronunciation accuracy. The participants in the study had mostly favorable attitudes towards AI-driven tools, emphasizing their effectiveness in enhancing pronunciation skills, boosting confidence, and promoting engagement. Nevertheless, several obstacles pertaining to the interpretation of feedback and the capture of subtle differences in pronunciation were recognized. This research has shown how AI can potentially be used to teach pronunciation effectively, and it has provided valuable insights for teachers, curriculum developers, and learners.

Articulatory characteristics and vowel space analysis of Mandarin Chinese non-low vowels in Korean-speaking learners

Vowels are a fundamental component of Mandarin Chinese syllables. Accurate pronunciation of Mandarin Chinese relies on the distinct articulation of different vowels, especially the visually similar “ü” and “u”. Korean-speaking learners of Chinese Mandarin as a second language commonly exhibit non-standard pronunciation or errors in articulating “ü”, leading to confusion, particularly in non-low vowels centered around this sound. Using Praat to analyze the Formant values of non-low vowel sounds from six participants, a comparison between Korean L2 Chinese speakers and native Chinese speakers revealed vowel spaces differentiated through F2 values for /u/ and /o/, while /ɣ/ and /o/ are distinguished by F1 values. The sound similar to Korean “ㅟ” differs from /y/ in that the former is a gliding sound transitioning from low to high. The distinction between /y/ and /u/, /o/, /ɣ/ was found to be unclear, with /y/ often merging into sounds like “iu” and “io”. This aligns with historical vowel changes in Chinese “ü”, noted in previous linguistic studies, showing variations like “iu” and “io” over different periods. Korean L2 Chinese speakers tend to substitute /ɯ/ for /u/ and “ㅟ” (/ɯ/+/i/) for /y/, avoiding rounded lip movements in pronunciation, thus reducing effort and the need for additional phonemes. Furthermore, the F3 value for Korean L2 Chinese /y/ is notably lower, suggesting that the correct articulation should closely resemble the tongue position of /i/, combined with rounded lips.

Error Analysis of English Pronunciation Acquisition by Chinese Non-English Majors

Chinese non-English majors are a large group of English learners. In the process of English pronunciationacquisition, issues such as incomplete phonological knowledge, transfer of mother tongue, and overgeneralization, lead to confusion of phonemes and stress, misunderstanding of syllable structure, and errors of assimilation, drop, and epenthesis. The accuracy of English pronunciation can only be improved by knowing both English and Chinese phonological systems, strengthening the teaching of English phonological knowledge, and adopting various phonological training activities.

Speech recognition sensors and artificial intelligence automatic evaluation application in English oral correction system

In China, although many college students have achieved good results in English exams, their oral proficiency is relatively weak. With the rapid development of internet information technology, the use of speech recognition technology to improve English oral pronunciation learning has become a research hotspot. The goal of this article is to design a system based on speech recognition and speech synthesis to help learners improve their oral pronunciation. By automatically evaluating learners' pronunciation accuracy and providing targeted correction and training, the system aims to improve learners' pronunciation level. This article adopts the principles and algorithms of speech recognition and speech synthesis. Through speech recognition technology, the system can convert learners' oral recordings into text and evaluate accuracy. By utilizing speech synthesis technology, the system can generate corrected audio based on correction rules to assist learners in pronunciation training. By evaluating and correcting real learners' oral recordings, the system can accurately evaluate learners' pronunciation accuracy and generate targeted corrected audio. Through systematic training and practice, learners' pronunciation level has been effectively improved.

Teachers’ Perceptions on the Role and Challenges of using ICT in English Language Classrooms

This paper assessed teachers’ perceived role and limitations on using ICT as a pedagogical tool in enhancing English language fluency. The study employed a descriptive case study design in which interviews, focus group discussions and document analysis were used to collect data. The collected data were analyzed qualitatively through recording the data for thematic procedures. It was found that teachers perceived ICT as an important pedagogical tool for enhancing grammar, vocabulary and accurate pronunciation. However, the use of ICT for teaching English language was limited due to poor ICT infrastructures, insufficient training, teachers’ readiness and unclear policies. The study recommends the government to make a critical investment in ICT facilities and ensure high internet connectivity in secondary schools. A practical training for English teachers on the adoption and integration of ICT through virtual platforms would certainly improve the practice.

Students’ Experience in Using Natural Reader Application to Ensure the Accuracy of Their English Pronunciation

Since 1960, speech synthesis technology has existed and its applications are now widely available, one of which is the Natural Reader application. Natural Reader is an application that can be used in knowing how to pronounce words in English. This study aims to describe students' experiences with using the Natural Reader application and examine the effect of utilizing the use of the Natural Reader application as a tool to test the accuracy of pronunciation of Mahmud Yunus Batusangkar State Islamic University students, by using the Natural Reader application students can more easily test their English accuracy. The method used by researchers is quantitative method, quantitative research can show data containing numbers that can be obtained by utilizing google form as a means of making a questionnaire. This research shows that the Natural Reader application has the potential to ensure the accuracy of pronunciation in English. So it can be concluded that the Natural Reader device can ensure the accuracy of student pronunciation in English.

Improve English Learning through Artificial Intelligence for Online and Offline Mixed Teaching Path

With the development of higher education in China, English has been listed as a compulsory course in basic education, and people’s demand for English teaching level is also increasing. However, with the advent of the information age, there have been significant changes in the dissemination of knowledge, teaching methods, and teaching environment. Traditional English teaching models can no longer meet the requirements of modern society for versatile talents. Therefore, exploring the mixed online and offline English teaching path has become particularly important. Artificial intelligence (AI) is an advanced technology today. It can evaluate students’ pronunciation and provide corresponding guidance, thereby increasing students’ interest in English and improving English pronunciation accuracy. It can also provide personalized learning content and learning paths according to students' learning needs and levels. By analyzing students' learning situation and feedback, it automatically adjusts and optimizes learning resources so that students can better master English knowledge and skills. The traditional English teaching model is often limited by time and space, while the AI-based hybrid teaching model can integrate a variety of educational resources around the world, including high-quality course content, learning resources and online tutoring services, so that students can get the most suitable educational resources anytime and anywhere, which can improve learning efficiency.

Improving speaking skill through hypno-teaching in Islamic boarding school

This study investigates the potential of e-portfolios to enhance vocabulary mastery and speaking ability in non-English as Foreign Language (EFL) students. Specifically, it explores the impact of e-portfolios on vocabulary acquisition, speaking fluency, and pronunciation accuracy in a non-EFL setting. The study employs a mixed-methods approach. A quantitative design with pre- and post-intervention tests measures vocabulary knowledge and speaking skills. A qualitative strand uses interviews and document analysis to explore students' perceptions and experiences with e-portfolios. Sixty non-EFL students are randomly assigned to an experimental group using e-portfolios and a control group receiving traditional instruction. Quantitative data is analysed using statistical tests to compare pre- and post-intervention performance between groups. Qualitative data is thematically analysed to identify emergent themes related to students' experiences with e-portfolios. The study expects to demonstrate that e-portfolios can effectively enhance vocabulary mastery and speaking ability in non-EFL students. Findings will contribute to the understanding of e-portfolio implementation in non-EFL contexts and inform pedagogical practices for improving vocabulary speaking skills. This study stands out by investigating the efficacy of e-portfolios in a non-EFL setting, addressing a gap in existing research. Additionally, the mixed-methods approach provides a comprehensive understanding of the impact of e-portfolios on both learning outcomes and learner experiences.

A pedagogical note on teaching L2 prosody and speech sounds using hand gestures

Abstract This paper reports the pedagogical decisions in Li et al. (2023) “The effectiveness of embodied prosodic training in L2 accentedness and vowel accuracy” published in Second Language Research. The study revealed that embodied prosodic training improved the overall pronunciation skills of Catalan-speaking learners of French and the pronunciation accuracy of front rounded vowels more than training without hand gestures. Three main pedagogical innovations were key in obtaining the positive results. First, the hand gestures performed by the instructor closely mimicked the prosodic structure (i.e., the melodic and rhythmic features) of the target sentences. Second, the sentences used for training were embedded in meaningful communicative dialogues. Third, the challenging L2 sounds were embedded in multiple occasions and in various prosodic positions. Therefore, in the hope of encouraging L2 teachers to incorporate such techniques in teaching practice, we provide a full description of the materials and procedures in Li et al. (2023).

Pronunciation Anxiety, Pronunciation-Related Views and Pronunciation Learning Actions of EMI and English Major Students

The present study explored pronunciation views, learning actions and anxiety in two differentiated English learner groups: one group which was studying English in an engineering degree and another group enrolled in an English major degree. The results from a questionnaire and focus group sessions revealed that the groups share some views such as a wish to improve their English pronunciation and adherence to native pronunciation references as well as pronunciation learning actions such as watching series or talking to native speakers. However, we also observed significant differences between the groups as the English major students tended to view pronunciation learning as a discipline, while the engineering students saw it as a communicative skill. The English majors also exhibited more pronunciation anxiety, specifically fear of ridicule and worry of making mistakes. They also revealed more anxiety-related themes in the focus group discussions such as worry about their pronunciation, pronunciation shame or unease because they do not know how to pronounce a word or cannot understand. Finally, the English major group was also more demanding of lecturers’ accurate pronunciation. These results suggest that differences regarding English pronunciation anxiety between different learner profiles may need to be considered during their English language learning progress.

Non-Linguistic Sounds and the Early Linguists

This paper aims at examining the non-linguistic sounds as they were tackled by the earlier linguists. They were not directly concerned with this topic but they studied it randomly such as explaining an idea or citing an example about the organs of speech. Their neglect of this topic is due to their concern with phonological studies of the Holy Qur'an. Phonological studies, therefore, were mainly concerned with the Qur'anic readings and the accurate pronunciation of Arabic sounds as the Holy Qur'an was revealed to Prophet Mohammad. That is why phonological studies were not important for them because of their little benefit. However, the study of sounds in general recurred in linguistic books and books concerned with animals. Linguistic books relied mainly on watching and listening and then discerning the intention behind such sounds. This will be the concern of this paper which deals with sounds in general, their definition, and the difference between them and other linguistic sounds concerning their production and manner of articulation

Insights from Madrosatul Alsun: A Communicative Vocabulary Learning in Non-Formal Education | Nadzaraat min Madrasah Alsun: Ta'liim Al-Mufradaat Al-Ittishaaliy fiy Ta'liim Ghair Al-Rasmiyy

The use of communicative context, repetition, visual associations, and accurate pronunciation are essential elements contributing to Arabic vocabulary learning. This study investigates how communicative vocabulary learning and its contributing elements are implemented at Madrosatul Alsun in the context of non-formal education. The research method used is a qualitative approach with a case study. Data was collected through in-depth interviews, participant observation in the classroom, and documentation. The obtained data is qualitatively analyzed in three phases: first, through condensation involving data selection, purification, simplification, summarization, and transformation; second, through data display to facilitate understanding of the issues; and third, drawing conclusions, where data interpretation is performed. The results indicate that vocabulary learning is conducted by integrating it with other skills, using various techniques, including direct conjugation, definitions, synonyms, antonyms, and visual references, and translating only under specific conditions. The discussion of these findings provides new insights into the model of "meaningful words in meaningful learning," a pattern of vocabulary learning within a communicative approach in a course classroom context distinct from the formal education classroom context.

A Spatial-Temporal Graph Model for Pronunciation Feature Prediction of Chinese Poetry.

With the development of artificial intelligence, speech recognition and prediction have become one of the important research domains with wild applications, such as intelligent control, education, individual identification, and emotion analysis. Chinese poetry reading contains rich features of continuous pronunciations, such as mood, emotion, rhythm schemes, lyric reading, and artistic expression. Therefore, the prediction of the pronunciation characteristics of a Chinese poetry reading is the significance for the presentation of high-level machine intelligence and has the potential to create a high-level intelligent system for teaching children to read Tang poetry. Mel frequency cepstral coefficient (MFCC) is currently used to present important speech features. Due to the complexity and high degree of nonlinearity in poetry reading, however, there is a tough challenge facing accurate pronunciation feature prediction, that is, how to model complex spatial correlations and time dynamics, such as rhyme schemes. As for many current methods, they ignore the spatial and temporal characteristics in MFCC presentation. In addition, these methods are subjected to certain limitations on prediction for long-term performance. In order to solve these problems, we propose a novel spatial-temporal graph model (STGM-MHA) based on multihead attention for the purpose of pronunciation feature prediction of Chinese poetry. The STGM-MHA is designed using an encoder-decoder structure. The encoder compresses the data into a hidden space representation, while the decoder reconstructs the hidden space representation as output. In the model, a novel gated recurrent unit (GRU) module (AGRU) based on multihead attention is proposed to extract the spatial and temporal features of MFCC data effectively. The evaluation comparison of our proposed model versus state-of-the-art methods in six datasets reveals the clear advantage of the proposed model.

Digital Resonance: Enhancing Pronunciation through Tech-Infused Teaching Strategy

The intention of this research is to discover the implementation of Cake in teaching of pronunciation and to uncover the students’ perception toward the use of Cake in teaching of pronunciation. This research applied qualitative descriptive method. The study was conducted at a private vocational school in Cianjur West Java Indonesia in the academic year 2021. The participants consisted of second grade of banking major at the ages between 15-17. Two of them were male and ten were female. This research collected the data through observation and a forced-choice questionnaire format where the participants were asked to choose one of the available selections (Yes or No). As the data were collected, the writer analyzed them. There were six steps in analyzing the observation as well as the questionnaire: (1) preparing the data to be analyzed, (2) reading or viewing all data, (3) coding the data, (4) generalizing the code, (5) creating tables based on the data code, and (6) interpret the data code. Over the observation the writers found a new formula on implementing Cake in helping students’ pronunciation accuracy. This formula then is called MRCP with drilling as the teaching strategy. The results of questionnaire show that Cake was new for the student. It has many features that support students in learning pronunciation. The students were also happy learning using Cake and they felt more confident to speak English after learning using Cake.

Ethnic group affiliation and second/foreign language accentedness in English and Mandarin among Hong Kong speakers

ABSTRACT As part of a larger project that investigates the issue of identities in Hong Kong, this study anchored on the sociocognitive paradigm in second language acquisition (SLA) explores the potential relationship between one’s identity and perceived language accentedness. Our study set in Hong Kong (HK) aims to extend Gatbonton and colleagues’ works (e.g. [2005]. Learners’ ethnic group loyalty and L2 pronunciation accuracy: A sociolinguistic investigation. TESOL Quarterly, 39(3), 489–511. https://doi.org/10.2307/3588491; [2008]. The ethnic group affiliation and L2 proficiency link: Empirical evidence. Language Awareness, 17(3), 229–248. https://doi.org/10.1080/09658410802146867; [2011]. Ethnic group affiliation and patterns of development of a phonological variable. Modern Language Journal, 95(2), 188–204. https://doi.org/10.1111/j.1540-4781.2011.01177.x) that examine the relationship between ethnic group affiliation (EGA) and language proficiencies in diglossic contexts. HK is a multi-glossic context where Cantonese, English and Mandarin are the official languages, and they perform distinctive functions in various public and private domains. Through analysing participants’ (n = 65; born between 1970s–1990s) self-identification and their reported accentedness in English and Mandarin, we address the question of whether EGA as a set of social factors has a bearing on a person’s linguistic achievements. Findings indicate that participants’ identification with the Chinese/ HK identity is related to their perceived accentedness in the targeted languages in intricate ways that do not align completely with our predictions. We conclude by calling for further socio-cognitively informed research that investigates multiglossic situations where languages/ language varieties complement or compete with each other.

The Effects of Academic Anxiety on College Students Learning of Spoken English

This paper examines the effects of academic anxiety on college students learning of spoken English. Academic anxiety can hinder students fluency, pronunciation accuracy, self-expression, and overall confidence in oral communication. The study identifies various factors that contribute to academic anxiety, including personal, external, socio-cultural, and psychological factors. It explores the negative consequences of academic anxiety on students spoken English learning outcomes. Coping mechanisms and strategies, such as creating a supportive learning environment, teaching stress management techniques, and promoting self-reflection and self-assessment, are proposed to address academic anxiety. The paper emphasizes the practical implications for educators, institutions, and policymakers, including teacher training, curriculum design, peer interaction, counseling services, and assessment practices. Implementing these recommendations can create a supportive and empowering learning environment, mitigating the impact of academic anxiety. The study concludes by highlighting the importance of recognizing and addressing academic anxiety to enhance college students spoken English skills and prepare them for success in a globalized world.

AUTOMATIC PRONUNCIATION ASSESSMENT IN A FLIPPED CLASSROOM CONTEXT

Automatic pronunciation assessment programs allow learners to practice pronunciation independently and receive immediate feedback based on acoustic analysis of their speech compared to a pronunciation model. The programs display a high level of reliability when assessing speakers’ pronunciation accuracy. The paper aims to present the results of an investigation into a recent experimental application of an automatic pronunciation assessment online program used within the flipped classroom of English phonetics and phonology course taught to foreign language pre-service teachers of English. Flipped learning approach was selected to offload the preliminary activities and tasks that do not require the immediate attention of the teacher to an online learning management system. Thus, course participants’ face-to-face learning could be enhanced and more effective. The participants could practice and receive tailored feedback from an online program before the face-to-face class. After the individual pronunciation practice, they could discuss more complex questions with the lecturer and, in collaboration with fellow students. The paper will present the survey results among the course participants who evaluated their experience with automatic pronunciation evaluation. Special attention is paid to the potential use of such programs for developing linguistic competence.

Exploring the Perceptions of Undergraduate University Students about their English Speaking skills in Hazara Division

This study aimed to investigate the perception of BS students about their English speaking skills. The study’s objectives were, to measure English language speaking skills of BS students in Universities of Hazara Division Pakistan and to find out possible reasons of weak speaking skills of English language of BS students in Universities of Hazara Division Pakistan. The population of the study comprised of all enrolled BS students in English Department of two selected universities, Abbottabad University of Science and Tchnology and COMSATS Abbottabad , Pakistan. Data was collected from 50 undergraduate students of English Department of both universities by using systematic sampling technique through a questionnaire consisted of 40 items. For data analysis, percentage and frequency technique were used as statistical techniques. Results revealed that University students generally display confidence in speaking English, expressing ideas clearly, and utilizing vocabulary, but face challenges in maintaining conversations and understanding diverse accents. Pronunciation and grammar skills are considered strong, yet fear of speaking and limited exposure hinder progress. Addressing language balance in classes, enhancing motivation, promoting idiomatic usage, and improving comprehension strategies are key areas requiring focused intervention for students to excel in English language proficiency. It was recommended that Encourage exposure to native speaker audio materials. students listen to improve pronunciation, educators can recommend and provide access to authentic audio resources like podcasts, interviews, and recordings. This exposure can help students develop accurate pronunciation and intonation to improve there language skills.

Did I say ‘pup’ or ‘pub’?

Language users possess knowledge about the second language (L2) phonological system and their own L2 pronunciation, known as phonological awareness and phonological self-awareness, respectively. Higher awareness about the L2 phonology has been shown to be beneficial for L2 pronunciation accuracy (e.g. Kennedy & Trofimovich, 2010), making it relevant to know which features L2 learners are able to notice and how the L2 phonology can be brought into the learners’ attention better. The aim of the paper is to examine language learners’ awareness about their own pronunciation by investigating which L2 segmental features are noticed, how one’s intelligibility is perceived and how participants view their phonological self-awareness abilities. 33 L1 Finnish learners of English enrolled in an English phonetics and phonology course recorded a speech sample and later listened to it by noting down the perceived pronunciation deviations on selected features. The participants also answered a phonological self-awareness questionnaire. The results indicate that the learners noticed segmental deviations in their productions, mainly in relation to consonantal voicing and vowel duration. The participants also perceived themselves as highly intelligible and reported it to be easy to notice the gap in their pronunciation, to identify pronunciation deviations in other speakers’ speech and to identify Finnish-accented English. Nevertheless, the overall verbalization of noticing was scarce and the participants manifested difficulties in explaining the noticed phenomena. The discussion relates the phonological self-awareness findings to pedagogical applications.