Pushing the Limits of Semi-Supervised Learning for Automatic Speech Recognition

10/20/2020

∙

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 test/test-other sets against the current state-of-the-art WERs 1.7

READ FULL TEXT

Pushing the Limits of Semi-Supervised Learning for Automatic Speech Recognition

Improved Noisy Student Training for Automatic Speech Recognition

Fully Learnable Front-End for Multi-Channel Acoustic Modeling using Semi-Supervised Learning

SESQA: semi-supervised learning for speech quality assessment

Semi-Supervised Noisy Student Pre-training on EfficientNet Architectures for Plant Pathology Classification

Likelihood-based semi-supervised model selection with applications to speech processing

Semi-supervised Learning with Sparse Autoencoders in Phone Classification

KinSPEAK: Improving speech recognition for Kinyarwanda via semi-supervised learning methods

Pushing the Limits of Semi-Supervised Learning for Automatic Speech Recognition

Related Research

Improved Noisy Student Training for Automatic Speech Recognition

Fully Learnable Front-End for Multi-Channel Acoustic Modeling using Semi-Supervised Learning

SESQA: semi-supervised learning for speech quality assessment

Semi-Supervised Noisy Student Pre-training on EfficientNet Architectures for Plant Pathology Classification

Likelihood-based semi-supervised model selection with applications to speech processing

Semi-supervised Learning with Sparse Autoencoders in Phone Classification

KinSPEAK: Improving speech recognition for Kinyarwanda via semi-supervised learning methods