Multilingual and code-switching ASR challenges for low resource Indian languages

TL;DR

This paper addresses the challenges of developing multilingual and code-switching automatic speech recognition systems for seven Indian languages, providing data, baselines, and analysis of low-resource scenarios.

Contribution

It introduces a new dataset and baseline models for multilingual and code-switching ASR involving seven Indian languages, focusing on low-resource conditions.

Findings

Baseline WER of 30.73% for multilingual ASR

Baseline WER of 32.45% for code-switching ASR

Provides insights into low-resource language ASR challenges

Abstract

Recently, there is 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 labeled 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 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 challenge, we would like to focus on building multilingual and code-switching 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 a baseline recipe for both the tasks with a WER of 30.73% and 32.45% on the test sets of multilingual and code-switching subtasks, respectively.