Self-Training and Pre-Training are Complementary for Speech Recognition
2021pp. 3030–3034
Citations Over TimeTop 1% of 2021 papers
Qiantong Xu, Alexei Baevski, Tatiana Likhomanenko, Paden Tomasello, Alexis Conneau, Ronan Collobert, Gabriel Synnaeve, Michael Auli
Abstract
Self-training and unsupervised pre-training have emerged as effective approaches to improve speech recognition systems using unlabeled data. However, it is not clear whether they learn similar patterns or if they can be effectively combined. In this paper, we show that pseudo-labeling and pre-training with wav2vec 2.0 are complementary in a variety of labeled data setups. Using just 10 minutes of labeled data from Libri-light as well as 53k hours of unlabeled data from LibriVox achieves word error rates (WER) of 2.8%/4.8% on the clean and other test sets of Librispeech – rivaling the best published systems trained on 960 hours of labeled data only a year ago. Training on all labeled data of Librispeech achieves WERs of 1.5%/3.1%.
Related Papers
- → Self-Training with Selection-by-Rejection(2012)19 cited
- → Automatic Labeling of Tweets for Crisis Response Using Distant Supervision(2020)13 cited
- → Evaluating Modeling Units and Sub-word Features in Language Models for Turkish ASR(2018)3 cited
- → Automatic selection of transcribed training material(2005)12 cited
- → Enhancing Self-Training Methods(2023)