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%.
%0 Generic
%1 zhang2020pushing
%A Zhang, Yu
%A Qin, James
%A Park, Daniel S.
%A Han, Wei
%A Chiu, Chung-Cheng
%A Pang, Ruoming
%A Le, Quoc V.
%A Wu, Yonghui
%D 2020
%K semisup speech
%T Pushing the Limits of Semi-Supervised Learning for Automatic Speech
Recognition
%U http://arxiv.org/abs/2010.10504
%X 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%.
@misc{zhang2020pushing,
abstract = {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%.},
added-at = {2020-11-18T10:52:27.000+0100},
author = {Zhang, Yu and Qin, James and Park, Daniel S. and Han, Wei and Chiu, Chung-Cheng and Pang, Ruoming and Le, Quoc V. and Wu, Yonghui},
biburl = {https://www.bibsonomy.org/bibtex/23cf7bea0e920a094f988247f10cd41f1/topel},
interhash = {cffd99bb0bf48859b9d9f2d8e42dbf3d},
intrahash = {3cf7bea0e920a094f988247f10cd41f1},
keywords = {semisup speech},
note = {cite arxiv:2010.10504Comment: 11 pages, 3 figures, 5 tables. Submitted to NeurIPS SAS 2020 Workshop},
timestamp = {2020-11-18T10:52:27.000+0100},
title = {Pushing the Limits of Semi-Supervised Learning for Automatic Speech
Recognition},
url = {http://arxiv.org/abs/2010.10504},
year = 2020
}