Yu Zhang, James Qin, Daniel S. Park, Wei Han, Chung-Cheng Chiu, Ruoming Pang, Quoc V. Le, Yonghui Wu
We employ a combination of recent developments in semi-supervised learning for automatic speech recognition to obtain state-of-the-art results on LibriSpeech utilizing the unlabeled audio of the Libri-Light dataset. More precisely, we carry out noisy student training with SpecAugment using giant Conformer models pre-trained using wav2vec 2.0 pre-training. By doing so, we are able to achieve word-error-rates (WERs) 1.4%/2.6% on the LibriSpeech test/test-other sets against the current state-of-the-art WERs 1.7%/3.3%.
| Task | Dataset | Metric | Value | Model |
|---|---|---|---|---|
| Speech Recognition | LibriSpeech test-clean | Word Error Rate (WER) | 1.4 | Conformer + Wav2vec 2.0 + SpecAugment-based Noisy Student Training with Libri-Light |
| Speech Recognition | LibriSpeech test-other | Word Error Rate (WER) | 2.6 | Conformer + Wav2vec 2.0 + SpecAugment-based Noisy Student Training with Libri-Light |