Discriminative Self-training for Punctuation Prediction
This work addresses the need for improved readability and downstream NLP performance in ASR outputs, though it is incremental as it builds on existing self-training methods.
The paper tackles the problem of punctuation prediction for automatic speech recognition transcripts by proposing a Discriminative Self-Training approach, which achieves a 1.3% absolute F1 gain over the current state-of-the-art on the IWSLT2011 test set.
Punctuation prediction for automatic speech recognition (ASR) output transcripts plays a crucial role for improving the readability of the ASR transcripts and for improving the performance of downstream natural language processing applications. However, achieving good performance on punctuation prediction often requires large amounts of labeled speech transcripts, which is expensive and laborious. In this paper, we propose a Discriminative Self-Training approach with weighted loss and discriminative label smoothing to exploit unlabeled speech transcripts. Experimental results on the English IWSLT2011 benchmark test set and an internal Chinese spoken language dataset demonstrate that the proposed approach achieves significant improvement on punctuation prediction accuracy over strong baselines including BERT, RoBERTa, and ELECTRA models. The proposed Discriminative Self-Training approach outperforms the vanilla self-training approach. We establish a new state-of-the-art (SOTA) on the IWSLT2011 test set, outperforming the current SOTA model by 1.3% absolute gain on F$_1$.