Improved Noisy Student Training for Automatic Speech Recognition

19 May 2020 Daniel S. Park Yu Zhang Ye Jia Wei Han Chung-Cheng Chiu Bo Li Yonghui Wu Quoc V. Le

Recently, a semi-supervised learning method known as "noisy student training" has been shown to improve image classification performance of deep networks significantly. Noisy student training is an iterative self-training method that leverages augmentation to improve network performance... (read more)

PDF Abstract

Results from the Paper


Ranked #2 on Speech Recognition on LibriSpeech test-clean (using extra training data)

     Get a GitHub badge
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK USES EXTRA
TRAINING DATA
RESULT BENCHMARK
Speech Recognition LibriSpeech test-clean ContextNet + SpecAugment-based Noisy Student Training with Libri-Light Word Error Rate (WER) 1.7 # 2
Speech Recognition LibriSpeech test-other ContextNet + SpecAugment-based Noisy Student Training with Libri-Light Word Error Rate (WER) 3.4 # 3

Methods used in the Paper