End-to-End Speech Recognition of Tamil Language

Changrampadi, Mohamed H.; Shahina, A.; Narayanan, M. Badri; Khan, A. Nayeemulla

doi:10.32604/iasc.2022.022021

Cited by 19 publications

(6 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Analyzing phonemes at the lowest level is called the phonetic phase (18) . An algorithm for unit selection speech synthesis matches the Phoneme with the prerecorded speech during phonetic analysis (19) .…”

Section: Software Requirementsmentioning

confidence: 99%

Tamil Speech Synthesizer App for Android: Text Processing Module Enhancement

Arulprakash¹,

Synthiya²,

Vijila³

et al. 2023

IJST

View full text Add to dashboard Cite

Objectives: Designing dynamic computer systems that are effective, efficient, simple, and satisfying to use is becoming extremely important in this age of information and communication technology. Text to Speech or Speech Synthesis is one of the many methods being investigated by researchers to improve Human-Computer Interaction. The goal here is to improve the text processing component of the Tamil voice synthesizer by including a text normalizer and loan word identification that is efficient and reliable. Methods: Text normalization is conducted on unconstrained Tamil text to turn non-standard terms into common words to reduce confusing utterances during intermediate word processing. Loan/Native words in Tamil literature are detected to enhance the Tamil voice synthesizer system's pronunciation model. Findings: During normalization, non-standard Tamil words are replaced with standard ones to reduce ambiguous utterances during interim processing. A pronunciation model is built to improve the Tamil speech synthesizer system by identifying loan words in Tamil text. A syllable classifier is presented in this study, based on a decision list approach, which can handle various types of non-stationary sounds. Novelty: We also disclose a 'loan/native word classifier' based on multiple linear regressions that perform well even with small words of three syllables. Such sophisticated text processors are required in today's dominating Digital, Information-Communication Technology, and Human-Computer Interaction age.

show abstract

Section: Software Requirementsmentioning

confidence: 99%

Tamil Speech Synthesizer App for Android: Text Processing Module Enhancement

Arulprakash¹,

Synthiya²,

Vijila³

et al. 2023

IJST

View full text Add to dashboard Cite

show abstract

“…(2) Drawing on the idea of the forward algorithm of HMM, the dynamic programming algorithm is used to effectively calculate the CTC loss function and its derivative, thus solving the problem of end-to-end training of CNN [9].…”

Section: Core Idea Of Ctcmentioning

confidence: 99%

Speech Recognition via CTC-CNN Model

Sung,

Kang,

Hsiao

2023

Computers, Materials &Amp; Continua

View full text Add to dashboard Cite

In the speech recognition system, the acoustic model is an important underlying model, and its accuracy directly affects the performance of the entire system. This paper introduces the construction and training process of the acoustic model in detail and studies the Connectionist temporal classification (CTC) algorithm, which plays an important role in the end-to-end framework, established a convolutional neural network (CNN) combined with an acoustic model of Connectionist temporal classification to improve the accuracy of speech recognition. This study uses a sound sensor, ReSpeaker Mic Array v2.0.1, to convert the collected speech signals into text or corresponding speech signals to improve communication and reduce noise and hardware interference. The baseline acoustic model in this study faces challenges such as long training time, high error rate, and a certain degree of overfitting. The model is trained through continuous design and improvement of the relevant parameters of the acoustic model, and finally the performance is selected according to the evaluation index. Excellent model, which reduces the error rate to about 18%, thus improving the accuracy rate. Finally, comparative verification was carried out from the selection of acoustic feature parameters, the selection of modeling units, and the speaker's speech rate, which further verified the excellent performance of the CTCCNN_5 + BN + Residual model structure. In terms of experiments, to train and verify the CTC-CNN baseline acoustic model, this study uses THCHS-30 and ST-CMDS speech data sets as training data sets, and after 54 epochs of training, the word error rate of the acoustic model training set is 31%, the word error rate of the test set is stable at about 43%. This experiment also considers the surrounding environmental noise. Under the noise level of 80∼90 dB, the accuracy rate is 88.18%, which is the worst performance among all levels. In contrast, at 40-60 dB, the accuracy was as high as 97.33% due to less noise pollution.

show abstract

Section: Core Idea Of Ctcmentioning

confidence: 99%

Speech Recognition via CTC-CNN Model

Sung

Kang

Hsiao

2022

Preprint

View full text Add to dashboard Cite

In modern society, human communication is increasingly frequent, and speech plays an extremely important role in social activities. Words can be used to express emotions and thoughts. However, numerous individuals are troubled by "language barriers", due to which their communication with others is limited. This study proposes a method to address the needs of speech-impaired and deaf-mute individuals. A basic deep neural network (DNN) acoustic model was established through a voice database combined with a deep neural network. A sound sensor was used to convert the collected voice signals and process them into text or corresponding voice signals to improve communication. This method can be extended to modern artificial intelligence technology, with diversified applications, such as verbatim transcripts of meeting minutes, medical reports, automotive, sales, etc. The results obtained in this study demonstrate the efficiency of the proposed method and discuss its significance.

show abstract

End-to-End Speech Recognition of Tamil Language

Cited by 19 publications

References 19 publications

Tamil Speech Synthesizer App for Android: Text Processing Module Enhancement

Tamil Speech Synthesizer App for Android: Text Processing Module Enhancement

Speech Recognition via CTC-CNN Model

Speech Recognition via CTC-CNN Model

Contact Info

Product

Resources

About