Self-training and Pre-training are Complementary for Speech Recognition

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 WERs of 3.0%/5.2% 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%.

PDF Abstract

Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Uses Extra
Training Data
Result Benchmark
Speech Recognition LibriSpeech test-clean wav2vec_wav2letter Word Error Rate (WER) 2.7 # 31
Speech Recognition LibriSpeech test-clean Conv + Transformer + wav2vec2.0 + pseudo labeling Word Error Rate (WER) 1.5 # 3
Speech Recognition LibriSpeech test-other Conv + Transformer + wav2vec2.0 + pseudo labeling Word Error Rate (WER) 3.1 # 4
Speech Recognition LibriSpeech train-clean-100 test-clean wav2vec_wav2letter Word Error Rate (WER) 2.8 # 1
Speech Recognition LibriSpeech train-clean-100 test-other wav2vec_wav2letter Word Error Rate (WER) 3.6 # 1


No methods listed for this paper. Add relevant methods here