Phonetic Classification Research Articles

The role of vowel and consonant onsets in neural tracking of natural speech

Objective. To investigate how the auditory system processes natural speech, models have been created to relate the electroencephalography (EEG) signal of a person listening to speech to various representations of the speech. Mainly the speech envelope has been used, but also phonetic representations. We investigated to which degree of granularity phonetic representations can be related to the EEG signal. Approach. We used recorded EEG signals from 105 subjects while they listened to fairy tale stories. We utilized speech representations, including onset of any phone, vowel–consonant onsets, broad phonetic class (BPC) onsets, and narrow phonetic class onsets, and related them to EEG using forward modeling and match–mismatch tasks. In forward modeling, we used a linear model to predict EEG from speech representations. In the match–mismatch task, we trained a long short term memory based model to determine which of two candidate speech segments matches with a given EEG segment. Main results. Our results show that vowel–consonant onsets outperform onsets of any phone in both tasks, which suggests that neural tracking of the vowel vs. consonant exists in the EEG to some degree. We also observed that vowel (syllable nucleus) onsets exhibit a more consistent representation in EEG compared to syllable onsets. Significance. Finally, our findings suggest that neural tracking previously thought to be associated with BPCs might actually originate from vowel–consonant onsets rather than the differentiation between different phonetic classes.

Automatic speech recognition of Portuguese phonemes using neural networks ensemble

The automatic speech recognition based on detection of phonemes provides advantages for online recognition of a speech represented by a sound signal. The development of a system for automatic speech recognition is multidisciplinary. It covers several areas of research, such as linguistics, signal processing and computational intelligence. In this work, the process starts with a speech signal pre-processing to extract the main features of the speech signal at a given instant of time. Inspired by the “divide and conquer” principle, we bridge the complexity gap of automatic speech recognition by devising models based on an ensemble of neural network experts, allowing to divide the huge decision space regarding speech recognition so that each expert takes care only of a delimited area of this decision space. This novel application of this strategy improves the precision, sensitivity and accuracy of the recognition process. Each included expert decides regarding each one of the pre-processed input samples. The decision set thus obtained is weighted. So, the expert with the highest weight for the output will determine the sample final classification. After that, a dynamic post-processing step, implemented as a recurrent neural network, is executed. It aims at mitigating the oscillatory effect that occurs during the recognition of classes with similar characteristics. In this work, two ensembles are investigated. The first is based on the clustering of similar phonetics classes while the second takes care of the imbalanced distribution of samples in the training set. The proposed model achieves 7.63% improvement in terms of accuracy with respect to the best so far related model for automatic speech recognition.

Using PRAAT in Teaching the Monophthongs to the EFL Students in Indonesia

<p>PRAAT can easily and rapidly improve the students’ ability in pronouncing monophthongs in phonetics class. The effectiveness of this software can help the students acquire quality monophthongs sounds. Furthermore, this software could reduce the problem of students learning sounds of monophthongs accurately such as the accuracy of F1 and F2 that is contained in monophthongs. It provided many programs on how to record, analyze, measure, and save in the wave file. This research aimed to investigate to what extent this software affects teaching monophthongs sounds in phonetics class for students. This research was conducted with 40 samples of students. They divided the class into 20 students for the control group and another 20 students for the experimental group. This quasi-experimental was to determine the best treatment before and after giving the treatment by using CALL for the experimental group and NON-CALL for the control group. Before using the software, the students were given the pre-test . Finally, they sat for post- test in the last meeting. The result was exposed through analysis of a t-test. It showed a significant difference in learning through PRAAT software (p<.05). In addition to that, teaching vowels sounds proved to be more successful using PRAAT than the traditional method.</p>

Информационные технологии на занятиях по фонетике как средство стимулирования студентов к профессиональному самосовершенствованию

Языковую компетентность можно охарактеризовать в различных измерениях и аспектах, из которых наиболее очевидными являются речь, письмо, аудирование (восприятие на слух) и чтение. Аудирование и чтение основаны на мыслительных процессах интерпретации звуковых и графических языковых сигналов, воспринимаемых слуховыми и зрительными рецепторами. Большинство методистов оперируют терминами языковая и речевая компетентности и относят эти понятия к концептуальным основам обучения родному языку. Другие утверждают, что «основной целью в обучении языку является формирование и развитие языковой личности – человека, который проявляет высокий уровень языковой и речевой компетентности». В. Виноградов в 30-х годах ХХ в. представил термин «языковая личность» и подметил, что языковая личность «вместилище социально речевых форм и норм коллектива; слияние различных языковых категорий». На современном этапе понятие «языковая личность» исследуется гораздо глубже: как модель речевых поступков; семиологическая личность; полилектная (общечеловеческая) и идеолектная (присуща определенной общности) личность; словарная языковая личность; национально-языковая личность; активная языковая личность. Именно в таком ключе, занятия по фонетике обретают особую актуальность среди студентов.Language competence can be characterized in various dimensions and aspects, of which the most obvious are speech, writing, listening (listening perception) and reading. Listening and reading are based on the mental processes of interpreting sound and graphic language signals perceived by auditory and visual receptors. Most methodologists use the terms language and speech competence and refer these concepts to the conceptual foundations of teaching the native language. Others argue that "the main goal in language teaching is the formation and development of a language personality – a person who shows a high level of language and speech competence." V. Vinogradov in the 30s of the twentieth century introduced the term "language personality" and noticed that the language personality is "a receptacle of social speech forms and norms of the collective; merging different language categories". At the present stage, the concept of "linguistic personality" is being investigated much more deeply: as a model of speech actions; semiological personality; poly-intellectual (universal) and ideolect (inherent in a certain community) personality; dictionary language personality; national language personality; active language personality. It is in this vein that phonetics classes are becoming particularly relevant among students.

Dance and Language in Phonetics Class

The chapter focuses on specialized communication and, more exactly, the use of technical terms between laypersons and experts of dancing. The phonetics of German is regarded as a resource for conveying specialist knowledge via the speech organs activated in the articulation of sounds, vowel quantity and quality during pronunciation training. Using material from a series of workshops, it can be demonstrated how the learner - the contemporary dancer - relates the specialist knowledge from the field of German phonetics to the specialist knowledge about the qualities of movement, and how s/he includes body movement in the pronunciation training which becomes a performative practice.

Improving Speech Enhancement Performance by Leveraging Contextual Broad Phonetic Class Information

Improving Speech Enhancement Performance by Leveraging Contextual Broad Phonetic Class Information

The TUNJO battery as a predictor of phonetic ability: A survey among English Philology students

Some learners are more successful in foreign language mastering than others. Among the plausible explanations discussed in the literature (Carroll 1981; Skehan 1991; Dörnyei 2005; Stansfield & Reed 2019; Griffiths & Soruç 2020), the concept of foreign language aptitude (FLA) is regarded as one of the key factors that can influence or predict learners’ success in the process of foreign language acquisition. The present pilot quantitative study aims to assess the extent to which learners’ level of foreign language aptitude can be correlated to their general phonological ability based on the example of first-year MA English Philology students (N=10). To assess the students’ level of aptitude, the Polish adaptation of the Modern Language Aptitude Test (MLAT), called the Test of Aptitude for the Learning of Foreign Languages (Test Uzdolnień do Nauki Języków Obcych – TUNJO), was used. On the other hand, to measure their level of phonetic ability, the test, which focused on several chosen areas covered during practical and theoretical phonetics classes during the BA programme, was constructed and submitted to the group. The quantitative data gathered throughout those two stages were subsequently analysed and interpreted. The results obtained revealed no significant correlation between the students’ level of aptitude and their general phonetic ability. Other individual differences and affective factors in language learning, alongside the structure of the measuring tools and the measurement itself, may justify the apparent lack of correlation.

Effectiveness of Mobile Learning in Enhancing Saudi English Majors’ Skills in Learning Vowels

Learning Phonetics, particularly the transcription of the vowel sounds, is one of the challenges that non-native English majors face. It can be best learned using images and videos as supplementary materials in the phonetics class in addition to the teacher’s live demonstration. Mobile pronunciation applications encompass all these features and provide an affordable and accessible resource for learning. The present study investigates the effect of using a pronunciation application on non-native Saudi students majoring in English at Majmaah University. The study provides insights on how mobile learning can be an effective tool for both English instructors and learners in teaching and learning vowels. A Pretest–posttest design was adopted in the present case study and descriptive and inferential statistical analyses were applied. The results indicated that short vowels are more challenging to the learners than long vowels and diphthongs. Findings also revealed that there is a significant difference between the learners’ performance before and after using the application which indicated the positive effect of using it in developing non-native learners’ skills in learning Phonetics.

Teaching teachers phonetics: The design and implementation of an asynchronous online English phonetics course

It is axiomatic that one of the chief goals of an applied linguistics program is to instruct teachers in the intricacies of English language structure. Explicit knowledge of the target language can help domestic and international teachers when dealing with adult 2nd language learners. But while most programs offer courses in English grammar, we found a paucity of (online) phonetics classes. We discuss three characteristics to be included in an online phonetics course: the description and learning of the sounds of the world’s languages, the technology-based collaborative procedures to narrowly transcribe a wide range of accented English speech, and the specific design to engage a variety of online students. Particular attention is devoted to our unique collaborative online project that at once trains students in the phonetic analysis of non-native speech. The results of these analyses are contributed to the online database, the speech accent archive (accent.gmu.edu), thereby giving students ownership of a publicly available online archive. The outcomes are described, with justifications and specific methods for measuring them. This paper emphasizes that learning to narrowly transcribe leads to enhanced listening and analysis, and that peer-to-peer collaboration is vital for any asynchronous online class.

Automatic phonetic classification of vocalic allophones in Tol

The aim of the present study involving automatic phonetic classification of /e/ and /u/ tokens in Tol is two-fold: first, I test existing claims about allophonic variation within these vowel classes, and second, I investigate allophonic variation within these vowel classes that has yet to be documented. The acoustic phonetic classifications derived in the present study contribute to a more detailed understanding of the allophonic systems operating within the Tol language. Operationalizing machine learning algorithms to investigate under-resourced, indigenous languages has the potential to provide detailed insights into the acoustic phonetic dynamics of a diverse range of vocalic systems.

TalkAbroad: A Relaxed Setting for Learning and More Fun than Cheating

For French Phonetics, in-class work stresses oral practice and feedback, whereas the online component includes lecture-based material, online low-stakes testing, links to , and other resources. Students then create a login and pay a fee for conversations in advance (but we had curriculum development grants for the initial Francophonie and Phonetics classes). Students are very motivated in that they each get to choose a conversation partner after reading partner profiles (in French) listing their job, hobbies, favorite books, music, movies, etc. Student reflections brought home the fact that the young Quebecers on TalkAbroad are indeed les enfants de la loi 101, the 1977 bill making French the official language of the provincial government and highlighted the importance of linguistic legislation to preserve the culture and language of Quebec.

Vocalic Ratio as One of the Most Important Criteria of Phonetic Classification of World Languages

Introduction. The article analyses the vowel-consonant ratio as one of the most important criteria of phonetic typology in the world languages. Scientific relevance of the research is based on quantitative and qualitative analysis and comparison of grammar and phonetics in typologically, genetically and historically different languages.Methodology and sources. Certain language is determined by vocalic ratio – a concept introduced to identify the vowels-consonant relation and measured through vk = V/C. Thus, all the languages can be either vocalic (vk > 1.3), consonantal (vk < 0.7) or mixed (0.7 > vk > 1.3). The article concerns the ideas by Ferdinand de Saussure (Indo-European root’s structure) and Aleksander V. Isachenko (phonetic typology).Results and discussion. The author conducts a comparative analysis of phonological systems and phonetic analysis of text fragments in several languages of different families and different historical periods: Gothic, old English, old Icelandic, English, Danish, French, and Finnish. The research reveals how the language’s structure matches its vowel-consonant ratio, i. e. disclose a link between its phonetic and morphology-syntactic classifications.Conclusion. The research has proved the fact that analytic trends in phonemes, on the one hand, depend on the vowel-consonant distribution in the language and speech, and on historically determined difference between the phonemes’ function – on the other. Inevitably, too, the language’s evolution from inflectional-synthetic to analytic or agglutinative (analytic-agglutinative) type affects all language levels, including the phonetic one. Consonants are stronger and almost resistible to changes; they function to distinguish the sense, making relative words so similar. The development of vowel system triggers the development of analytic functions, which are bound to impact the language system. Increasing number of vowels, emerging diphthongs and triphthongs are the result of analytic abilities of the language.

Can we decode phonetic features in inner speech using surface electromyography?

Although having a long history of scrutiny in experimental psychology, it is still controversial whether wilful inner speech (covert speech) production is accompanied by specific activity in speech muscles. We present the results of a preregistered experiment looking at the electromyographic correlates of both overt speech and inner speech production of two phonetic classes of nonwords. An automatic classification approach was undertaken to discriminate between two articulatory features contained in nonwords uttered in both overt and covert speech. Although this approach led to reasonable accuracy rates during overt speech production, it failed to discriminate inner speech phonetic content based on surface electromyography signals. However, exploratory analyses conducted at the individual level revealed that it seemed possible to distinguish between rounded and spread nonwords covertly produced, in two participants. We discuss these results in relation to the existing literature and suggest alternative ways of testing the engagement of the speech motor system during wilful inner speech production.

Broad Phonetic Classification of ASR using Visual Based Features

Abstract: This paper presents a novel method of classifying speech phonemes. Four hybrid techniques based on the acoustic-phonetic approach and pattern recognition approach are used to emphasize the principle idea of this research. The first hybrid model is constructed of fixed state, structured Hidden Markov Model, Gaussian Mixture, Mel scaled Best Tree Image, Convolution Neural network, Vector Quantization (FS-HMM-GM-MBTI-CNN-VQ). The second hybrid model is constructed of variable state, dynamically structured Hidden Markov Model, Gaussian Mixture, Mel scaled Best Tree Image, Convolution Neural network, Vector Quantization (VS-HMM-GM-MBTI-CNN-VQ). The third hybrid model is constructed of fixed state, structured Hidden Markov Model, Gaussian Mixture, Mel scaled Best Tree Image, Convolution Neural network (FS-HMM-GM-MBTI-CNN). The fourth hybrid model is constructed of variable state, dynamically structured Hidden Markov Model, Gaussian Mixture, Mel scaled Best Tree Image, Convolution Neural network (VS-HMM-GM-MBTI-CNN). TIMIT database is used in this paper. All phones are classified into five classes and segregated into Vowels, Plosives, Fricatives, Nasals, and Silences. The results show that using (VS-HMM-GM-MBTI-CNN-VQ) is an available method for classification of phonemes, with the potential for use in applications such as automatic speech recognition and automatic language identification. Competitive results are achieved especially in nasals, plosives, and silence high successive rates than others.

ANFIS for Tamil Phoneme Classification

Phoneme recognition is an intricate problem lying under non-linear systems. Most research in this area revolve around try to model the pattern of features observed in the speech spectra with the use of Hidden Markov Models (HMM), various types of neural networks like deep recurrent neural networks, time delay neural networks, etc. for efficient phoneme recognition. In this paper, we study the effectiveness of the hybrid architecture, the Adaptive Neuro-Fuzzy Inference System (ANFIS) for capturing the spectral features of the speech signal to handle the problem of Phoneme Recognition. In spite of a wide range of research in this field, here we examine the power of ANFIS for least explored Tamil phoneme recognition problem. The experimental results have shown the ability of the model to learn the patterns associated with various phonetic classes, indicated with recognition improvement in terms of accuracy to its counterparts.

Phonetic acquisition in cortical dynamics, a computational approach.

Many computational theories have been developed to improve artificial phonetic classification performance from linguistic auditory streams. However, less attention has been given to psycholinguistic data and neurophysiological features recently found in cortical tissue. We focus on a context in which basic linguistic units–such as phonemes–are extracted and robustly classified by humans and other animals from complex acoustic streams in speech data. We are especially motivated by the fact that 8-month-old human infants can accomplish segmentation of words from fluent audio streams based exclusively on the statistical relationships between neighboring speech sounds without any kind of supervision. In this paper, we introduce a biologically inspired and fully unsupervised neurocomputational approach that incorporates key neurophysiological and anatomical cortical properties, including columnar organization, spontaneous micro-columnar formation, adaptation to contextual activations and Sparse Distributed Representations (SDRs) produced by means of partial N-Methyl-D-aspartic acid (NMDA) depolarization. Its feature abstraction capabilities show promising phonetic invariance and generalization attributes. Our model improves the performance of a Support Vector Machine (SVM) classifier for monosyllabic, disyllabic and trisyllabic word classification tasks in the presence of environmental disturbances such as white noise, reverberation, and pitch and voice variations. Furthermore, our approach emphasizes potential self-organizing cortical principles achieving improvement without any kind of optimization guidance which could minimize hypothetical loss functions by means of–for example–backpropagation. Thus, our computational model outperforms multiresolution spectro-temporal auditory feature representations using only the statistical sequential structure immerse in the phonotactic rules of the input stream.

Semi-supervised learning of a nonnative phonetic contrast: How much feedback is enough?

Semi-supervised learning refers to learning that occurs when feedback about performance is provided on only a subset of training trials. Algorithms for semi-supervised learning are popular in machine learning because of their minimal reliance on labeled data. There have been, however, only a few reports of semi-supervised learning in humans. Here we document human semi-supervised learning on a nonnative phonetic classification task. Classification performance remained unchanged when 60 feedback trials were provided on each of the two days of training. In contrast, performance improved when 60 feedback trials were combined with 240 no-feedback trials each day. In variants of this successful semi-supervised regimen, increasing the daily number of feedback trials from 60 to 240 did not increase the amount of learning, while decreasing that number to 30 abolished learning. Finally, replacing the no-feedback trials with stimulus exposure alone had little effect on the outcome. These results were an unexpected consequence of combining training periods with feedback and testing periods without feedback, illustrating that no-feedback testing can influence learning outcomes. More broadly, these data suggest that task performance with feedback can function as an all-or-none trigger for recruiting the contribution of trials without feedback, or mere stimulus exposures, to human learning.

The automatic assessment of the severity of dysphonia

Perceptual evaluation of the patient’s voice is the most commonly used method in everyday clinical practice. We propose an automatic approach for the prediction of severity of some types of organic and functional dysphonia. By means of an unsupervised learning method, we have demonstrated that acoustic parameters measured on different phonetic classes are suitable for modelling the four grade assessments of the specialists (RBH subjective scale from 0 to 3). In this study, the overall hoarseness H was examined. Four specialists were asked to determine the severity of dysphonia. A k-means cluster analysis was performed for the decision of each specialist separately; the average accuracy of the four-grade classification was 0.46. The four-grade classification has been surprisingly close to the subjective judgements. Moreover, automatic estimation of severity of dysphonia was also determined. Linear regression and RBF kernel regression models were compared. The average rating of the four specialists were used as target in the experiments. Low RMSE and high correlation measures were obtained between the automatically predicted severity and perceptual assessments. The best RMS value of H was 0.45 for the model with RBF kernel, however, a simpler linear model provided the highest correlation value of 0.85, using only eight acoustic parameters.

СОПОСТАВЛЕНИЕ РЕЗУЛЬТАТОВ ЛИНГВОСТАТИСТИЧЕСКОГО АНАЛИЗА ПЕРЦЕПТИВНЫХ ТИПОВ РУССКОГО И АНГЛИЙСКОГО СЛОВА (НА МАТЕРИАЛЕ НКРЯ, БНК, АНК)

Corpus research presents obvious benefits, though linguists approach the material in various ways. For example, corpus linguists approach data in an exploratory way, whereas psycholinguists more often tend to combine corpus data and experimental research. The current work uses the theoretical systemic approach to describe the two frequency strata of the three corpora (Russian National Corpus, British National Corpus and Open American National Corpus) and build the classification of phonetic word types in Russian and English (British and American). The aim of the research is to draw up the phonetic (perceptive) classification of the corresponding languages and to describe the identity of their sound systems based on these types. The high frequency and the frequency strata of the three corpora have been analyzed to identify the words characterized by the following linguistic features: length in syllables, stressed vowel, rhythmic structure, etc. The data comparison discovered more distinctions than similarities among the three corpora...

Relative occurrences and difference of extrema for detection of transitions between broad phonetic classes

Detection of transitions between broad phonetic classes in a speech signal has applications such as landmark detection and segmentation. The proposed hierarchical method detects silence to non-silence transitions, sonorant to non-sonorant transitions and vice-versa. The subset of the extrema (minimum or maximum amplitude samples) above a threshold, occurring between every pair of successive zero-crossings, is selected from each frame of the bandpass-filtered speech signal. Locations of the first and the last extrema lie on either side far away from the mid-point (reference) of a frame, if the speech signal belongs to a non-transition segment; else, one of these locations lies within a few samples from the reference, indicating a transition frame. The transitions are detected from the entire TIMIT database for clean speech and 93.6% of them are within a tolerance of 20 ms from the phone boundaries. Sonorant, unvoiced non-sonorant and silence classes and their respective onsets are detected with an accuracy of about 83.5% for the same tolerance with respect to the labelled TIMIT database as reference. The results are as good as, and in some aspects better than, the state-of-the-art methods for similar tasks. The proposed method is also tested on the test set of the TIMIT database for robustness with respect to white, babble and Schroeder noise, and about 90% of the transitions are detected within a tolerance of 20 ms at the signal to noise ratio of 5 dB. On NTIMIT database, 62.7% of the transitions are detected, and 63.5% of the sonorant onsets, within 20 ms tolerance.