Dual Script E2E framework for Multilingual and Code-Switching ASR

TL;DR

This paper introduces a dual script end-to-end framework for multilingual and code-switching ASR in Indian languages, leveraging a rule-based phoneme-level label set and novel back-end techniques to improve recognition accuracy.

Contribution

It proposes two innovative E2E ASR systems using a common label set and native script recovery, advancing multilingual and code-switching speech recognition for Indian languages.

Findings

Achieved approximately 6% WER reduction in multilingual ASR

Achieved approximately 5% WER reduction in code-switching ASR

Demonstrated effectiveness on the Indic ASR Challenge 2021

Abstract

India is home to multiple languages, and training automatic speech recognition (ASR) systems for languages is challenging. Over time, each language has adopted words from other languages, such as English, leading to code-mixing. Most Indian languages also have their own unique scripts, which poses a major limitation in training multilingual and code-switching ASR systems. Inspired by results in text-to-speech synthesis, in this work, we use an in-house rule-based phoneme-level common label set (CLS) representation to train multilingual and code-switching ASR for Indian languages. We propose two end-to-end (E2E) ASR systems. In the first system, the E2E model is trained on the CLS representation, and we use a novel data-driven back-end to recover the native language script. In the second system, we propose a modification to the E2E model, wherein the CLS representation and the native language characters are used simultaneously for training. We show our results on the multilingual and code-switching tasks of the Indic ASR Challenge 2021. Our best results achieve 6% and 5% improvement (approx) in word error rate over the baseline system for the multilingual and code-switching tasks, respectively, on the challenge development data.