Mucs 2021: Multilingual and code-switching asr challenges for low resource indian languages

Diwan, A and Vaideeswaran, R and Shah, S and Singh, A and Raghavan, S and Khare, S and Unni, V and Vyas, S and Rajpuria, A and Yarra, C and Mittal, A and Ghosh, PK and Jyothi, P and Bali, K and Seshadri, V and Sitaram, S and Bharadwaj, S and Nanavati, J and Nanavati, R and Sankaranarayanan, K (2021) Mucs 2021: Multilingual and code-switching asr challenges for low resource indian languages. In: 22nd Annual Conference of the International Speech Communication Association, INTERSPEECH 2021, 30 Aug - 03 Sep 2021, Brno, pp. 351-355.

PDF INTERSPEECH_2021.pdf - Published Version Restricted to Registered users only Download (169kB) | Request a copy

Abstract

Recently, there is an increasing interest in multilingual automatic speech recognition (ASR) where a speech recognition system caters to multiple low resource languages by taking advantage of low amounts of labelled corpora in multiple languages. With multilingualism becoming common in today's world, there has been increasing interest in code-switching ASR as well. In code-switching, multiple languages are freely interchanged within a single sentence or between sentences. The success of low-resource multilingual and code-switching (MUCS) ASR often depends on the variety of languages in terms of their acoustics, linguistic characteristics as well as the amount of data available and how these are carefully considered in building the ASR system. In this MUCS 2021 challenge, we would like to focus on building MUCS ASR systems through two different subtasks related to a total of seven Indian languages, namely Hindi, Marathi, Odia, Tamil, Telugu, Gujarati and Bengali. For this purpose, we provide a total of ₆₀₀ hours of transcribed speech data, comprising train and test sets, in these languages, including two code-switched language pairs, Hindi-English and Bengali-English. We also provide baseline recipes1 for both the subtasks with 30.73 and 32.45 word error rate on the MUCS test sets, respectively. Copyright © 2021 ISCA.

Item Type:	Conference Paper
Publication:	Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH
Publisher:	International Speech Communication Association
Additional Information:	The copyright for this article belongs to International Speech Communication Association
Department/Centre:	Division of Electrical Sciences > Electrical Engineering
Date Deposited:	03 Dec 2021 08:51
Last Modified:	03 Dec 2021 08:51
URI:	http://eprints.iisc.ac.in/id/eprint/70640

Actions (login required)

View Item