An efficient text augmentation approach for contextualized Mandarin speech recognition

TL;DR

This paper introduces a simple and efficient text augmentation method that leverages text-only data to improve Mandarin speech recognition, especially for rare words, with minimal additional computational cost.

Contribution

It proposes a novel text-augmentation technique using a codebook to enhance pre-trained ASR models without extensive data or computation.

Findings

Up to 30% relative CER reduction on rare words

15% relative CER reduction overall

Effective with minimal computational overhead

Abstract

Although contextualized automatic speech recognition (ASR) systems are commonly used to improve the recognition of uncommon words, their effectiveness is hindered by the inherent limitations of speech-text data availability. To address this challenge, our study proposes to leverage extensive text-only datasets and contextualize pre-trained ASR models using a straightforward text-augmentation (TA) technique, all while keeping computational costs minimal. In particular, to contextualize a pre-trained CIF-based ASR, we construct a codebook using limited speech-text data. By utilizing a simple codebook lookup process, we convert available text-only data into latent text embeddings. These embeddings then enhance the inputs for the contextualized ASR. Our experiments on diverse Mandarin test sets demonstrate that our TA approach significantly boosts recognition performance. The top-performing system shows relative CER improvements of up to 30% on rare words and 15% across all words in general.