Learning Fast Adaptation on Cross-Accented Speech Recognition
Abstract
Local dialects influence people to pronounce words of the same language differently from each other. The great variability and complex characteristics of accents creates a major challenge for training a robust and accent-agnostic automatic speech recognition (ASR) system. In this paper, we introduce a cross-accented English speech recognition task as a benchmark for measuring the ability of the model to adapt to unseen accents using the existing CommonVoice corpus. We also propose an accent-agnostic approach that extends the model-agnostic meta-learning (MAML) algorithm for fast adaptation to unseen accents. Our approach significantly outperforms joint training in both zero-shot, few-shot, and all-shot in the mixed-region and cross-region settings in terms of word error rate.
Cite
@article{arxiv.2003.01901,
title = {Learning Fast Adaptation on Cross-Accented Speech Recognition},
author = {Genta Indra Winata and Samuel Cahyawijaya and Zihan Liu and Zhaojiang Lin and Andrea Madotto and Peng Xu and Pascale Fung},
journal= {arXiv preprint arXiv:2003.01901},
year = {2020}
}
Comments
The first three authors contributed equally to this work