Design and development of a large vocabulary, continuous speech recognition system for Tamil

Madhavaraj, A and Ramakrishnan, AG (2018) Design and development of a large vocabulary, continuous speech recognition system for Tamil. In: 14th IEEE India Council International Conference, INDICON 2017, 15 December 2017 through 17 December 2017, Roorkee.

PDF IEEE_INDICON_2017.pdf - Published Version Restricted to Registered users only Download (293kB) | Request a copy

Abstract

This paper presents our work on building a large vocabulary continuous speech recognition system for Tamil using deep neural networks (DNN). Well known techniques, namely, maximum likelihood linear transformation and speaker-adaptive training have been used to build our final deep neural network based speech recognition system. We have used 6.5 hours of Tamil speech recorded from 30 speakers covering a vocabulary of 13,026 words, of which 4.5 hours of data was used for training, 1 hour of data for testing and 1 hour of data for cross-validation. Two independent recognition systems were built, one for phone recognition (PR) and the other for continuous speech recognition (CSR) and they achieve phone error rate of 24.9 and word error rate of 3.5, respectively. DNN-based triphone acoustic model shows an absolute improvement of about 1 and 23 over the monophone acoustic model for CSR and PR, respectively. © 2017 IEEE.

Item Type:	Conference Paper
Publication:	2017 14th IEEE India Council International Conference, INDICON 2017
Publisher:	Institute of Electrical and Electronics Engineers Inc.
Additional Information:	The copyright for this article belongs to the Institute of Electrical and Electronics Engineers Inc.
Keywords:	Continuous speech recognition; Hidden Markov models; Linear transformations; Mathematical transformations; Maximum likelihood; Speech; Speech recognition; Telephone sets; Vocabulary control, Acoustic model; Design and Development; Language model; Large vocabulary continuous speech recognition; Recognition systems; Speaker adaptive trainings; Speech recognition systems; Tamil, Deep neural networks
Department/Centre:	Division of Electrical Sciences > Electrical Engineering
Date Deposited:	02 Aug 2022 05:56
Last Modified:	02 Aug 2022 05:56
URI:	https://eprints.iisc.ac.in/id/eprint/75212

Actions (login required)

View Item