Aiming at reducing the reliance on expensive human annotations, data synthesis for Automatic Speech Recognition (ASR) has remained an active area of research. While prior work mainly focuses on synthetic speech generation for ASR data augmentation, its combination with text generation methods is considerably less explored. In this work, we explore text augmentation for ASR using large-scale pre-trained neural networks, and systematically compare those to traditional text augmentation methods. The generated synthetic texts are then converted to synthetic speech using a text-to-speech (TTS) system and added to the ASR training data. In experiments conducted on three datasets, we find that neural models achieve 9%-15% relative WER improvement and outperform traditional methods. We conclude that text augmentation, particularly through modern neural approaches, is a viable tool for improving the accuracy of ASR systems.
@article{arxiv.2305.16333,
title = {Text Generation with Speech Synthesis for ASR Data Augmentation},
author = {Zhuangqun Huang and Gil Keren and Ziran Jiang and Shashank Jain and David Goss-Grubbs and Nelson Cheng and Farnaz Abtahi and Duc Le and David Zhang and Antony D'Avirro and Ethan Campbell-Taylor and Jessie Salas and Irina-Elena Veliche and Xi Chen},
journal= {arXiv preprint arXiv:2305.16333},
year = {2023}
}