Simple yet Effective Code-Switching Language Identification with Multitask Pre-Training and Transfer Learning

TL;DR

This paper introduces novel multitask pre-training and transfer learning methods, including a stacked Residual CNN+GRU model, to improve language identification in low-resource English-Mandarin code-switching speech, significantly outperforming baselines.

Contribution

It proposes a multitask pre-training approach with auxiliary ASR tasks and data augmentation techniques for effective CSLID in low-resource settings.

Findings

Achieved a balanced accuracy of 0.781 on English-Mandarin CSLID.

Outperformed previous baseline by 55.3%.

Demonstrated effectiveness across low-resource code-switching data.

Abstract

Code-switching, also called code-mixing, is the linguistics phenomenon where in casual settings, multilingual speakers mix words from different languages in one utterance. Due to its spontaneous nature, code-switching is extremely low-resource, which makes it a challenging problem for language and speech processing tasks. In such contexts, Code-Switching Language Identification (CSLID) becomes a difficult but necessary task if we want to maximally leverage existing monolingual tools for other tasks. In this work, we propose two novel approaches toward improving language identification accuracy on an English-Mandarin child-directed speech dataset. Our methods include a stacked Residual CNN+GRU model and a multitask pre-training approach to use Automatic Speech Recognition (ASR) as an auxiliary task for CSLID. Due to the low-resource nature of code-switching, we also employ careful silver data creation using monolingual corpora in both languages and up-sampling as data augmentation. We focus on English-Mandarin code-switched data, but our method works on any language pair. Our best model achieves a balanced accuracy of 0.781 on a real English-Mandarin code-switching child-directed speech corpus and outperforms the previous baseline by 55.3%.