scholarly journals Accent Recognition with Hybrid Phonetic Features

Sensors ◽  
2021 ◽  
Vol 21 (18) ◽  
pp. 6258
Author(s):  
Zhan Zhang ◽  
Yuehai Wang ◽  
Jianyi Yang
Keyword(s):  
Hybrid Structure ◽  
Acoustic Model ◽  
Acoustic Feature ◽  
Accented Speech ◽  
Controlled Systems ◽  
Language Knowledge ◽  
Accent Recognition ◽  
Phonetic Features ◽  
High Level ◽  
Abstract Feature

The performance of voice-controlled systems is usually influenced by accented speech. To make these systems more robust, frontend accent recognition (AR) technologies have received increased attention in recent years. As accent is a high-level abstract feature that has a profound relationship with language knowledge, AR is more challenging than other language-agnostic audio classification tasks. In this paper, we use an auxiliary automatic speech recognition (ASR) task to extract language-related phonetic features. Furthermore, we propose a hybrid structure that incorporates the embeddings of both a fixed acoustic model and a trainable acoustic model, making the language-related acoustic feature more robust. We conduct several experiments on the AESRC dataset. The results demonstrate that our approach can obtain an 8.02% relative improvement compared with the Transformer baseline, showing the merits of the proposed method.

Chinese-dominant bilingualism

2019 ◽  
Vol 42 (2) ◽  
pp. 280-322
Author(s):  
Hongdi Ding
Keyword(s):  
Native Speakers ◽  
Young Generation ◽  
Speech Community ◽  
Subtle Change ◽  
Quantitative Evidence ◽  
Language Knowledge ◽  
Aged Subjects ◽  
Current Sample ◽  
High Level ◽  
Almost All

Abstract This research provides quantitative evidence of the decline in Nuosu competence among the young Nuosu generation in Liangshan, Sichuan, China, through a direct comprehensive linguistic measurement of their Nuosu-Chinese bilingual competence. Although the young generation can still speak Nuosu, a Tibeto-Burman language, as fluently as the elder Nuosu generations without apparent difficulty, this research identifies the subtle change of competence before it becomes widely noticeable. A sample of 34 ethnic Nuosu of three generations was tested in Xichang, Liangshan, through measuring their core or implicit language knowledge (i.e. morphology, syntax, lexicon, semantics, and pragmatics) in Nuosu and Chinese. The participants were from seven Shynra-speaking counties and two Yynuo-speaking counties, mainly within Liangshan. The test format was listening and speaking, to include illiterate speakers. It was found that all elder and middle-aged subjects still possessed monolingual baseline competence, which was stable and maintained at a high level. However, only half of the young Nuosu subjects achieved monolingual baseline competence in Nuosu. The other half, though still considered as native speakers of Nuosu, had lower and more varied competence; however, almost all of them achieved monolingual baseline competence in Chinese. The Nuosu speech community in Liangshan is shifting from Nuosu-dominant bilingualism to Chinese-dominant bilingualism. Moreover, the present study proposes a typology of native speakers and a typology of bilinguals based on different levels of competence obtained from the current sample.

scholarly journals Template-Assisted Electrochemical Synthesis of CdSe Quantum Dots—Polypyrrole Composite Nanorods

2020 ◽  
Vol 10 (17) ◽  
pp. 5966
Author(s):  
Won-Seok Kang ◽  
Taegon Oh ◽  
Gwang-Hyeon Nam ◽  
Hyo-Sop Kim ◽  
Ki-Suk Kim ◽  
...  
Keyword(s):  
Simple Procedure ◽  
Hybrid Structure ◽  
Proof Of Concept ◽  
Interface Area ◽  
Anodic Aluminum ◽  
Enhanced Photoluminescence ◽  
Polypyrrole Composite ◽  
High Level ◽  
Oxide Membranes ◽  
Composite Nanorods

Luminescent nanoparticles have reached a high level of maturity in materials and spectral tunability for optics and optoelectronics. However, the lack of facile methodology for heterojunction formation of the nanoparticles provides many challenges for scalability. In this paper we demonstrate a simple procedure to synthesize a nanoparticle-embedded polymer nanorod hybrid structure via a template-based electrochemical method using anodic aluminum oxide membranes. This method enables the formation of interactive nanostructures wherein the interface area between the two components is maximized. As a proof of concept, semiconducting CdSe nanoparticles were embedded in polypyrrole nanorods with dimensions that can be finely tuned. We observed enhanced photoluminescence of the hybrid structures compared with bare polypyrrole nanorods.

Notion of an accent in linguistic studies

2021 ◽  
Vol IX(257) (75) ◽  
pp. 54-56
Author(s):  
N. O. Petrochuk
Keyword(s):  
Linguistic Features ◽  
Distinctive Features ◽  
Accented Speech ◽  
Structural Organisation ◽  
Phonetic Features ◽  
The Given

The given article introduces the main areas of studying an accent. Particular attention is given to the field of linguistics, phonetics, and phonological research where an accent is not only a characteristic of an individual but also a bearer of distinctive features of the foreign speech. These features include differences on various language levels such as phonological, morphological, lexical, syntactical. The linguistic and non-linguistic phonetic features are illustrated. The peculiarities in pronunciation, which include melodic arrangements of utterances, rhythmical and structural organisation of the sentence, pausation, articulations in addition to vowels' and consonants' production and their interaction in speech are described as related to linguistic features. Non-linguistic features are connected with the personality of a speaker, the listener, the situation of speech and the context. The article presents a short outline of the criteria to measure a foreign-accented speech.

scholarly journals Romanian Language Voice Browsing for Web Applications Using Grapheme Level Acoustic Modeling

2013 ◽  
Vol 8-9 ◽  
pp. 29-36 ◽  
Author(s):  
József Domokos ◽  
László Sándor ◽  
Ovidiu Buza ◽  
Gavril Toderean
Keyword(s):  
Web Application ◽  
Web Applications ◽  
Operating Mode ◽  
Acoustic Model ◽  
Continuous Speech Recognition ◽  
Application Development ◽  
Acoustic Feature ◽  
Vector Construction ◽  
Speech Database ◽  
Model Training

The aim of this article is to present a demonstrative Web application with Romanian language continuous speech recognition based multimodal interface. The scope of the paper also includes the presentation and testing of the capabilities of a context dependent grapheme based acoustic model for the Romanian language. The article describes the system architecture, the Web application development and the speech database used for the acoustic feature vector construction and acoustic model training. Further the task grammar is presented. At the end recognition results are presented in both offline and online operating mode. The used speech corpora together with the transcriptions are freely available for academic use on the NaviRo project website: http://users.utcluj.ro/~jdomokos/naviro/.

scholarly journals Universal competence of intercultural interaction in learning a foreign language in professional activities

2021 ◽  
Vol 127 ◽  
pp. 02006
Author(s):  
Maria Sergeevna Zavyalova ◽  
Elina Borisovna Kalinichenko ◽  
Lyubov Mikhailovna Ivanova ◽  
Marina Nikolaevna Razdobarova
Keyword(s):  
Foreign Language ◽  
Language Teachers ◽  
Professional Competencies ◽  
Educational Institutions ◽  
Higher Educational Institutions ◽  
Professional Activities ◽  
Intercultural Interaction ◽  
Language Knowledge ◽  
Higher Educational ◽  
High Level

The article is devoted to the problem of formation of universal competencies. The purpose of the work is to determine the optimal conditions for efficient formation of universal competencies of intercultural interaction. The authors substantiate the idea that, along with professional competencies, the employers increasingly demand a high level of universal competencies in modern graduates of higher educational institutions. The article dwells on the peculiarities of formation of intercultural-interaction universal competence as a complex characteristic of the graduate’s aptitude to apply the acquired foreign-language knowledge, skills and abilities in standard and professional situations. The research identified the interconnection between the universal professional and general professional competencies. The article considers the subject-oriented methods of teaching a foreign language, which along with the universal competence, contribute to the formation of students’ GPC (general professional competencies) and PC (professional competencies). The basic concepts in respect of the above are presented and analysed; the ratio of maturity of this competence in the process of foreign language teaching is highlighted. A number of methods necessary for the formation of intercultural-interaction universal competence were identified for the development of the educational/methodological complex under the discipline “Foreign language in professional activities” within the system of training for the master degree in the field 38.04.01 “Economics”. The considered subject will be interesting for specialists in different spheres of humanitarian knowledge: foreign language teachers of higher educational institutions, representatives of educational and methodical associations, methodologists.

scholarly journals Strategies of Speech Interaction between Adults and Preschool Children with Typical and Atypical Development

2019 ◽  
Vol 9 (12) ◽  
pp. 159 ◽  
Author(s):  
Elena Lyakso ◽  
Olga Frolova ◽  
Aleksey Grigorev ◽  
Viktor Gorodnyi ◽  
Aleksandr Nikolaev
Keyword(s):  
Developmental Disorders ◽  
Autism Spectrum ◽  
Speech Development ◽  
Typically Developing ◽  
Behavior Strategies ◽  
Speech Interaction ◽  
Phonetic Features ◽  
Atypical Development ◽  
High Level ◽  
Children's Speech

The goal of this research is to study the speech strategies of adults’ interactions with 4–7-year-old children. The participants are “mother–child” dyads with typically developing (TD, n = 40) children, children with autism spectrum disorders (ASDs, n = 20), Down syndrome (DS, n = 10), and “experimenter–orphan” pairs (n = 20). Spectrographic, linguistic, phonetic, and perceptual analyses (n = 465 listeners) of children’s speech and mothers’ speech (MS) are executed. The analysis of audio records by listeners (n = 10) and the elements of nonverbal behavior on the basis of video records by experts (n = 5) are made. Differences in the speech behavior strategies of mothers during interactions with TD children, children with ASD, and children with DS are revealed. The different strategies of “mother–child” interactions depending on the severity of the child’s developmental disorders and the child’s age are described. The same features of MS addressed to TD children with low levels of speech formation are used in MS directed to children with atypical development. The acoustic features of MS correlated with a high level of TD child speech development do not lead to a similar correlation in dyads with ASD and DS children. The perceptual and phonetic features of the speech of children of all groups are described.

scholarly journals Professionally-oriented foreign language training: problems and solutions

2021 ◽  
Author(s):  
Т.А. Танцура
Keyword(s):  
Foreign Language ◽  
Language Education ◽  
Foreign Language Education ◽  
Professional Communication ◽  
The European Union ◽  
Education Abroad ◽  
Language Knowledge ◽  
Problems And Solutions ◽  
High Level ◽  
The University

В статье рассматриваются основные направления иноязычного образования в вузе в настоящее время. Автор указывает на общую тенденцию трансформации иноязычного образования за рубежом и в отечественном образовании. Создание единого европейского государства и открытие границ способствовали росту мобильности профессиональных кадров, что вызвало потребность в высоком уровне знания иностранного языка для повседневного и профессионального общения. Актуальность статьи заключается в изучении проблемы внедрения интегрированного обучения профильных дисциплин посредством иностранного языка в вузе. Автором отмечается тот факт, что несмотря на возможность активизации процесса обучения профильным дисциплинам и одновременного совершенствования знаний и навыков иностранного языка, в настоящее время лишь небольшое количество вузов внедряют интегрированное обучение. В статье определяются трудности и преимущества введения обучения на основе предметно-языкового интегрированного обучения. The article deals with the main directions of foreign language education at the university at the present time. The author points to the general trend regarding to transformation of foreign language education abroad and in Russia. The creation of the European union and the opening of borders contributed to the growth of the mobility of professional personnel, which caused the need for a high level of a foreign language knowledge for everyday and professional communication. The relevance of the article is to study the problem of implementing integrated teaching of specialized disciplines through a foreign language at a university. The author notes the fact that despite the possibility of activating the process of teaching specialized disciplines and simultaneously improving the knowledge and skills of a foreign language, only a small number of universities are implementing integrated training currently. The article defines the difficulties and advantages of introducing training based on subject-language integrated learning.

scholarly journals Accent Recognition for Indian English using Acoustic Feature Approach

2013 ◽  
Vol 63 (7) ◽  
pp. 25-32 ◽  
Author(s):  
Santosh Gaikwad ◽  
Bharti Gawali ◽  
K. V. Kale
Keyword(s):  
Acoustic Feature ◽  
Indian English ◽  
Accent Recognition

scholarly journals Automatic Voice Query Service for Multi-Accented Mandarin Speech

2021 ◽  
Author(s):  
Kejing Xiao ◽  
Zhaopeng Qian
Keyword(s):  
Automatic Speech Recognition ◽  
Acoustic Model ◽  
Acoustic Features ◽  
Service Requirement ◽  
Matching Algorithm ◽  
Accented Speech ◽  
Real Scene ◽  
Fuzzy Mathematic ◽  
Fusion Feature ◽  
New Framework

Automatic Voice Query Service can extremely reduce the artificial cost, which could improve the response efficiency for users. The automatic speech recognition (ASR) is one of the important component in AVQS. However, many dialect areas in China make the AVQS have to response the multi-accented Mandarin users by single acoustic model in ASR. This problem severely limits the accuracy of ASR for multi-accented speech in the AVQS. In this paper, a new framework for AVQS is proposed to improve the accuracy of response. Firstly, the fusion feature including iVector and filterbank acoustic features is used to train the Transformer-CTC model. Secondly, the transformer-CTC model is used to construct the end-to-end ASR. Finally, key words matching algorithm for AVQS based on fuzzy mathematic theory is proposed to further improve the accuracy of response. The results show that the final accuracy in our proposed framework for AVQS arrives at 91.5%. The proposed framework for AVQS can satisfy the service requirement of different areas in mainland of China. This research has a great significance for exploring the application value of artificial intelligence in the real scene.

Sign in / Sign up

Export Citation Format

Share Document