Abstract
The two key components of Automatic Speech Recognition (ASR) and Text-to-Speech (TTS) systems are language modeling and acoustic modeling. The language model generates a lexicon, which is a pronunciation dictionary. A lexicon can be created using a variety of approaches. For low-resource languages, rule-based methods are typically employed to build the lexicon. However, because the corpus is often tiny, this methodology does not account for all possible pronunciation variances. As a result, low-resource languages like Malayalam require a method for developing a comprehensive lexicon as the corpus grows. In this work, we explored deep learning based encoder-decoder models for grapheme-to-phoneme (G2P) conversion in Malayalam. Long Short-Term Memory (LSTM) and Bidirectional Long Short-Term Memory (BiLSTM) encoder models with varying embedding dimensions were used to create the encoder model. The performance of the deep learning models used for G2P conversion was measured using the Word Error Rate (WER) and Phoneme Error Rate (PER). With 1024 embedding dimensions, the encoder using the BiLSTM model had the maximum accuracy of 98.04% and the lowest PER of 2.57% at the phoneme level, and the highest accuracy of 90.58% and the lowest WER of 9.42% at the word level.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Babu, L.B., George, A., Sreelakshmi, K.R., Mary, L.: Continuous speech recognition system for malayalam language using Kaldi. In: 2018 International Conference on Emerging Trends and Innovations In Engineering And Technological Research. ICETIETR, pp. 1–4. IEEE, Ernakulam (2018). https://doi.org/10.1109/ICETIETR.2018.8529045
Young, S.: The HTK hidden Markov model toolkit: Design and philosophy. University of Cambridge, Department of Engineering Cambridge, England (1994)
Sri, K.V.L., Srinivasan, M., Nair, R.R., Priya, K.J., Gupta, D.: Kaldi recipe in Hindi for word level recognition and phoneme level transcription. In: Procedia Computer Science, pp. 2476–2485 (2020)
Hannun, A., Case, C., Casper, J., Catanzaro, B., Diamos, G., Elsen, E., Prenger, R., Satheesh, S., Sengupta, S., Coates, A., Andrew Y. Ng.: Deep speech: scaling up end-to-end speech recognition. arXiv preprint arXiv:1412.5567 (2014)
Kumar, C.S., Govind, D., Nijil, C., Manish, N.: Grapheme to phone conversion for Hindi. In: Oriental COCOSDA. Malaysia (2006)
Chen, S.F.: Conditional and joint models for grapheme-to-phoneme conversion. In: EUROSPEECH-2003, pp. 2033–2036. Geneva (2003)
Taylor, P.: Hidden Markov models for grapheme to phoneme conversion. In: INTERSPEECH-2005, pp. 1973–1976. Lisbon (2005)
Rao, K., Peng, F., Sak, H., Beaufays, F.: Grapheme-to-phoneme conversion using long short-term memory recurrent neural networks. In: 2015 IEEE International Conference on Acoustics, Speech and Signal Processing. ICASSP, pp. 4225–4229. IEEE, South Brisbane (2015). https://doi.org/10.1109/ICASSP.2015.7178767
Yolchuyeva, S., Németh, G., Gyires-Tóth, B.: Grapheme-to-phoneme conversion with convolutional neural networks. Appl. Sci. 96, 1143 (2019)
Yolchuyeva, S., Németh, G., Gyires-Tóth, B.: Transformer based grapheme-to-phoneme conversion. In: INTERSPEECH-2019, ISCA (2019). https://doi.org/10.21437/interspeech.2019-1954
Premjith, B., Kumar, M.A., Soman, K.P.: Neural machine translation system for English to Indian language translation using MTIL parallel corpus. J. Intel. Syst. 28(3), 387–398 (2019)
Premjith, B., Soman, K.P., Poornachandran, P.: A deep learning based Part-of-Speech (POS) tagger for Sanskrit language by embedding character level features. In: Proceedings of the 10th annual meeting of the Forum for Information Retrieval Evaluation, pp. 56–60. Association for Computing Machinery, New York, USA (2018)
Krishnan, V.V., Anto, P.B.: Features of wavelet packet decomposition and discrete wavelet transform for Malayalam speech recognition. Int. J. Recent Trends Eng. 1(2), 93–96 (2009)
Lekshmi, K.R., Jithesh, V.S., Sherly, E.: Malayalam speech corpus: design and development for dravidian language. In: Proceedings of the WILDRE5–5th Workshop on Indian Language Data: Resources and Evaluation, pp. 25–28. European Language Resources Association (ELRA), Marseille (2020)
Nair, S.S., Rechitha, C.R., Kumar, C.S.: Rule-based grapheme to phoneme converter for Malayalam. Int. J. Comput. Linguist. Nat. Lang. Process. 2(7), 417–420 (2013)
Baby, A., Nishanthi, N.L., Thomas, A.L., Murthy, H.A.: A unified parser for developing Indian language text to speech synthesizers. In: International Conference on Text, Speech, and Dialogue. TSD 2016, LNCS, vol. 9924, pp. 514–521. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-45510-5_59
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Priyamvada, R., Govind, D., Menon, V.K., Premjith, B., Soman, K.P. (2022). Grapheme to Phoneme Conversion for Malayalam Speech Using Encoder-Decoder Architecture. In: Satapathy, S.C., Peer, P., Tang, J., Bhateja, V., Ghosh, A. (eds) Intelligent Data Engineering and Analytics. Smart Innovation, Systems and Technologies, vol 266. Springer, Singapore. https://doi.org/10.1007/978-981-16-6624-7_5
Download citation
DOI: https://doi.org/10.1007/978-981-16-6624-7_5
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-16-6623-0
Online ISBN: 978-981-16-6624-7
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)