Constructing a phonetic transcribed text corpus for Southern Thai dialect Speech Recognition

Aunkaew, Sittichok; Karnjanadecha, Montri; Wutiwiwatchai, Chai

doi:10.1109/jcsse.2015.7219772

Cited by 6 publications

(4 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Currently, in the field of dialect digitization research, F. de Vriend, L. Boves, et al have use Geographic Information Systems (GIS) to analyze and interpret dialect data, making visualization a tool for dialect research [20]. S. Aunkaew, M. Karnjanadecha and C. Wutiwiwatchai have proposed a hybrid technique to recognize the southern Thai dialect, constructing a continuous speech recognition corpus for the southern Thai dialect [21]. On the social status level of dialects, M. Krause and C. Sappok have analyzed the distribution of Russian dialect and the current status of relevant information differentiation and provided insights and prospects on the use of Russian dialects [22].…”

Section: B Relevant Studies and Optimizationmentioning

confidence: 99%

Research on Dialect Protection: Interaction Design of Chinese Dialects Based on BLSTM-CRF and FBM Theories

Han,

Zhu,

Wen

et al. 2024

IEEE Access

View full text Add to dashboard Cite

This study aims to augment the impact of dialect-related cultural elements among adolescents, thereby facilitating a more effective inheritance and progression of Chinese dialect culture within the contemporary socio-cultural milieu.Firstly, based on the LSTM architecture, the bidirectional long shortterm memory-conditional random field (BLSTM-CRF) model framework for dialect recognition was optimized and established. Subsequently, employing the BLSTM-CRF model for dialect recognition and the acquisition of dialectal voice samples, the natural language sequence input underwent conversion into word vectors or character vectors. The word vectors or character vectors were input into the BLSTM layer to obtain the hidden state vector corresponding to each position, representing the context information of that position. The hidden state vector was input into a fully connected layer, and the CRF layer was adopted to calculate and find the highest scoring label sequence as the final prediction result. Secondly, the FBM theoretical model was utilized to analyze the target user needs through user research methods. Finally, a design framework oriented by dialect recognition and user needs was constructed, and interactive product design was carried out.On the basis of BLSTM-CRF, the product design framework is improved to complete the interactive application design.Integrating user needs and Long Short-Term Memory (LSTM) algorithms into the design of dialect products, we explore ways to preserve and disseminate dialect culture among the younger generation. This provides new insights for the digital inheritance and development of dialects, and also serves as a reference for the design process of related cultural apps.

show abstract

Section: B Relevant Studies and Optimizationmentioning

confidence: 99%

Research on Dialect Protection: Interaction Design of Chinese Dialects Based on BLSTM-CRF and FBM Theories

Han,

Zhu,

Wen

et al. 2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…The forms of the syllables are: (1) Consonant + Vowel (CV) such as [ปา /pa:/ 'to hurl'] or [ดู / du:/ 'to look'], (2) Consonant + Vowel + Consonant (CVC) such as [บาน /ba:n/ 'to bloom'] or [กิ น /kin/ 'to eat']. A diphthong can be written as C(C) such as [ปลา /pla:/ 'fish'] or [คลาน /khla:n/ 'to crawl'] [33,34]. Basic Thai vowels are divided into two types: short vowels and long vowels.…”

Section: Thai Vowels and Pronunciation Problemsmentioning

confidence: 99%

“…V is a short vowel, such as [ติ /ti/ 'to censure'] or [ดุ /du/ 'to scold'], and V(V) is a long vowel, such as [ตี /ti:/ 'to hit'] or [ดู /du:/ 'to watch'] [35]. [33,34]. Basic Thai vowels are divided into two types: short vowels and long vowels.…”

Section: Thai Vowels and Pronunciation Problemsmentioning

confidence: 99%

See 1 more Smart Citation

An Acoustic Feature-Based Deep Learning Model for Automatic Thai Vowel Pronunciation Recognition

Rukwong

Pongpinigpinyo

2022

Applied Sciences

View full text Add to dashboard Cite

For Thai vowel pronunciation, it is very important to know that when mispronunciation occurs, the meanings of words change completely. Thus, effective and standardized practice is essential to pronouncing words correctly as a native speaker. Since the COVID-19 pandemic, online learning has become increasingly popular. For example, an online pronunciation application system was introduced that has virtual teachers and an intelligent process of evaluating students that is similar to standardized training by a teacher in a real classroom. This research presents an online automatic computer-assisted pronunciation training (CAPT) using deep learning to recognize Thai vowels in speech. The automatic CAPT is developed to solve the inadequacy of instruction specialists and the complex vowel teaching process. It is a unique system that develops computer techniques integrated with linguistic theory. The deep learning model is the most significant part of recognizing vowels pronounced for the automatic CAPT. The major challenge in Thai vowel recognition is the correct identification of Thai vowels when spoken in real-world situations. A convolutional neural network (CNN), a deep learning model, is applied and developed in the classification of pronounced Thai vowels. A new dataset for Thai vowels was designed, collected, and examined by linguists. The result of an optimal CNN model with Mel spectrogram (MS) achieves the highest accuracy of 98.61%, compared with Mel frequency cepstral coefficients (MFCC) with the baseline long short-term memory (LSTM) model and MS with the baseline LSTM model have an accuracy of 94.44% and 90.00% respectively.

show abstract

Thai-Dialect: Low Resource Thai Dialectal Speech to Text Corpora

Suwanbandit,

Chitiyaphol,

Chuenchom

et al. 2023

2023 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)

View full text Add to dashboard Cite

Constructing a phonetic transcribed text corpus for Southern Thai dialect Speech Recognition

Cited by 6 publications

References 5 publications

Research on Dialect Protection: Interaction Design of Chinese Dialects Based on BLSTM-CRF and FBM Theories

Research on Dialect Protection: Interaction Design of Chinese Dialects Based on BLSTM-CRF and FBM Theories

An Acoustic Feature-Based Deep Learning Model for Automatic Thai Vowel Pronunciation Recognition

Thai-Dialect: Low Resource Thai Dialectal Speech to Text Corpora

Contact Info

Product

Resources

About