Language Model Bootstrapping Using Neural Machine Translation For Conversational Speech Recognition

TL;DR

This paper proposes using neural machine translation for data augmentation to bootstrap conversational speech recognition models in new languages, demonstrating significant WER reduction especially in underrepresented interaction scenarios.

Contribution

It introduces domain adaptation techniques for effective MT-based data augmentation in speech recognition, addressing challenges like domain mismatch and named entities.

Findings

Achieved 7.8-15.6% relative WER reduction using MT-based augmentation.

Translation particularly improves underrepresented interaction scenarios.

Domain adaptation techniques enhance the effectiveness of neural machine translation for speech data.

Abstract

Building conversational speech recognition systems for new languages is constrained by the availability of utterances that capture user-device interactions. Data collection is both expensive and limited by the speed of manual transcription. In order to address this, we advocate the use of neural machine translation as a data augmentation technique for bootstrapping language models. Machine translation (MT) offers a systematic way of incorporating collections from mature, resource-rich conversational systems that may be available for a different language. However, ingesting raw translations from a general purpose MT system may not be effective owing to the presence of named entities, intra sentential code-switching and the domain mismatch between the conversational data being translated and the parallel text used for MT training. To circumvent this, we explore the following domain adaptation techniques: (a) sentence embedding based data selection for MT training, (b) model finetuning, and (c) rescoring and filtering translated hypotheses. Using Hindi as the experimental testbed, we translate US English utterances to supplement the transcribed collections. We observe a relative word error rate reduction of 7.8-15.6%, depending on the bootstrapping phase. Fine grained analysis reveals that translation particularly aids the interaction scenarios which are underrepresented in the transcribed data.