Pseudo Label Is Better Than Human Label

Conference of the International Speech Communication Association (INTERSPEECH)(2022)

引用 3|浏览17
暂无评分
摘要
State-of-the-art automatic speech recognition (ASR) systems are trained with tens of thousands of hours of labeled speech data. Human transcription is expensive and time consuming. Factors such as the quality and consistency of the transcription can greatly affect the performance of the ASR models trained with these data. In this paper, we show that we can train a strong teacher model to produce high quality pseudo labels by utilizing recent self-supervised and semi-supervised learning techniques. Specifically, we use JUST (Joint Unsupervised/Supervised Training) and iterative noisy student teacher training to train a 600 million parameter bi-directional teacher model. This model achieved 4.0% word error rate (WER) on a voice search task, 11.1% relatively better than a baseline. We further show that by using this strong teacher model to generate high-quality pseudo labels for training, we can achieve 13.6% relative WER reduction (5.9% to 5.1%) for a streaming model compared to using human labels.
更多
查看译文
关键词
speech recognition, pseudo label, semi-supervised, self-supervised
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要