Self-training and Pre-training are Complementary for Speech Recognition
arXiv:2010.11430
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 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%.
References in corpus (16)
- SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition
- Categorical Reparameterization with Gumbel-Softmax
- Libri-Light: A Benchmark for ASR with Limited or No Supervision
- Conformer: Convolution-augmented Transformer for Speech Recognition
- vq-wav2vec: Self-Supervised Learning of Discrete Speech Representations
- Improved Noisy Student Training for Automatic Speech Recognition
- Self-Training for End-to-End Speech Recognition
- wav2letter++: The Fastest Open-source Speech Recognition System
- End-to-end ASR: from Supervised to Semi-Supervised Learning with Modern Architectures
- The challenge of realistic music generation: modelling raw audio at scale
- Improving Transformer-based Speech Recognition Using Unsupervised Pre-training
- Effectiveness of self-supervised pre-training for speech recognition
- ContextNet: Improving Convolutional Neural Networks for Automatic Speech Recognition with Global Context
- Self-training Improves Pre-training for Natural Language Understanding
- Semi-Supervised Speech Recognition via Local Prior Matching
- Iterative Pseudo-Labeling for Speech Recognition
Cited by in corpus (5)
- Self-Training: A Survey
- Overcoming Domain Mismatch in Low Resource Sequence-to-Sequence ASR Models using Hybrid Generated Pseudotranscripts
- Comparison of Self-Supervised Speech Pre-Training Methods on Flemish Dutch
- DAP: Detection-Aware Pre-training with Weak Supervision
- Improving Streaming Transformer Based ASR Under a Framework of Self-supervised Learning