Identification of regional dialects of Telugu language using text independent speech processing models

Shivaprasad, S.; Sadanandam, M.

doi:10.1007/s10772-020-09678-y

Cited by 24 publications

(10 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition to this, these studies were primarily concentrated on English speakers (72.7%), while only 27.3% of studies involved other languages. A total number of 15 languages were explored, namely, French ( Hesling et al, 2005 ; Girard et al, 2008 ; Burki et al, 2011 ; Kennedy and Blanchet, 2014 ), Korean ( Mitterer et al, 2013 ; Kim et al, 2022 ), Greek ( Kambanaros, 2014 ), Mitterer and McQueen, 2009 ), Dutch ( Ernestus et al, 2017 ), Norwegian ( Kirmess and Lind, 2011 ), Telugu ( Hivaprasad and Sadanandam, 2020 ), Cantonese ( Yiu et al, 2002 ), Persian ( Daneshi et al, 2020 ), Finnish ( Alexandrou et al, 2017 ), Bengali ( Bose et al, 2022 ), Spanish ( Guzman et al, 2021 ; Gonzalez-Alvarez and Sos-Pena, 2022 ; Lofgren and Hinzen, 2022 ), Portuguese ( Brinca et al, 2014 ; Sampaio et al, 2019 ), Swedish ( Alves et al, 2020 ; Strombergsson et al, 2020 ), Mandarin ( Tsai et al, 2012 ), and Italian (e.g., Cerrato et al, 1998 ; Leoni and Cutugno, 1999 ). In addition to English, studies on Italian connected speech were more abundant than that of other languages.…”

Section: Resultsmentioning

confidence: 99%

A systematic review of studies on connected speech processing: Trends, key findings, and implications

Zare

Kania

et al. 2022

Front. Psychol.

View full text Add to dashboard Cite

Connected speech processing (CSP) is of great significance to individuals’ language and cognitive development. It is particularly crucial not only for clinical detection and treatment of developmental disorders, but also for the Foreign/second language teaching instructions. However, given the importance of this field, there is a clear lack of systematic reviews that summarize the key findings of previous studies. To this end, through searching in the scientific databases PsycInfo, Scopus, PubMed, ERIC, Taylor and Francis, and Web of Science, the present study identified 128 core CSP articles with high reference values according to PRISMA guidance and the following results were obtained through quantitative analysis and qualitative comparative synthesis: (1) The number of studies on CSP published per year showed an upward trend; however, most focused on English language, whereas the studies on other languages were comparatively rare; (2) CSP was found to be affected by multiple factors, among which speech speed, semantics, word frequency, and phonological awareness were most frequently investigated; (3) the deficit in CSP capacity was widely recognized as a significant predictor and indicator of developmental disorders; (4) more studies were carried out on connected speech production than on perception; and (5) almost no longitudinal studies have ever been conducted among either native or non-native speakers. Therefore, future research is needed to explore the developmental trajectory of CSP skills of typically developing language learners and speakers with cognitive disorders over different periods of time. It is also necessary to deepen the understanding of the processing mechanism beyond their performance and the role played by phonological awareness and lexical representations in CSP.

show abstract

Section: Resultsmentioning

confidence: 99%

A systematic review of studies on connected speech processing: Trends, key findings, and implications

Zare

Kania

et al. 2022

Front. Psychol.

View full text Add to dashboard Cite

show abstract

“…3 There is also some influence of the neighboring states' languages like Kannada on Telangana. The Coastal Andhra dialect is largely influenced by Sanskrit as well as Tamil due to historical and geographical reasons (Shivaprasad and Sadanandam, 2020). Finally, the Rayalaseema dialect is influenced by neighboring states' languages, i.e., Tamil and Kannada (Shivaprasad and Sadanandam, 2020).…”

Section: Methodsmentioning

confidence: 99%

Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies: Student Research Workshop

2022

View full text Add to dashboard Cite

Systematicity is thought to be a key inductive bias possessed by humans that is lacking in standard natural language processing systems such as those utilizing transformers. In this work, we investigate the extent to which the failure of transformers on systematic generalization tests can be attributed to a lack of linguistic abstraction in its attention mechanism. We develop a novel modification to the transformer by implementing two separate input streams: a role stream controls the attention distributions (i.e., queries and keys) at each layer, and a filler stream determines the values. Our results show that when abstract role labels are assigned to input sequences and provided to the role stream, systematic generalization is improved. ReferencesMarco Baroni. 2020. Linguistic generalization and compositionality in modern artificial neural networks.

show abstract

“…In Malaysian Quranic Dialect recognition 5.5 to 7Dialect recognition researches have been done in many Indian languages. Shivaprasad and Sadanandam [16] used GMM and HMM to identify regional Telugu dialects. The authors generated a database of Telugu dialects for this purpose.…”

Section: Literature Reviewmentioning

confidence: 99%

Assamese Dialect Identification System Using Deep Learning

Das

Bhattacharjee

2022

Preprint

View full text Add to dashboard Cite

The goal of a dialect Identification is to label speech in an audio file with dialect labels. This paper presents a method for automatically identifying four Assamese dialects: Central Assamese, Eastern Assamese dialect, Kamrupi dialect, and Goaplari dialect, using Convolution Neu-ral Networks (CNN). In this study, utterances of four major regional dialects of the Assamese language, namely Central Assamese spoken in and around Nagaon district, Eastern Assamese dialect spoken in the Sibsagar and its neighboring districts, Kamrupi dialect spoken in Kam-rup, Nalbari, Barpeta, Kokarajhar and some parts of Bongaigaon district and Goaplari dialect spoken in the Goaplara, Dhuburi and part of Bon-gaigaon district were used. The classifier was trained on audio samples from each of the four dialects that lasted 2 hours. The CNN uses Mel spectrogram images created from two to four seconds divisions of raw audio input with varied audio quality. The performance of the system is also examined as a function of train and test audio sample durations. When compared to machine learning models, the suggested CNN model obtains an accuracy of 90.82 percent, which may be considered the best.

show abstract

Identification of regional dialects of Telugu language using text independent speech processing models

Cited by 24 publications

References 15 publications

A systematic review of studies on connected speech processing: Trends, key findings, and implications

A systematic review of studies on connected speech processing: Trends, key findings, and implications

Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies: Student Research Workshop

Assamese Dialect Identification System Using Deep Learning

Contact Info

Product

Resources

About