Multimodal Semi-supervised Learning Framework for Punctuation Prediction in Conversational Speech
This work addresses the problem of improving punctuation accuracy in speech processing for applications like transcription and dialogue systems, representing an incremental advance with specific performance gains.
The paper tackles punctuation prediction in conversational speech by proposing a multimodal semi-supervised learning framework, achieving absolute F1 score improvements of ~6-9% on reference transcripts and ~3-4% on ASR outputs over a baseline BLSTM model, with further gains from data augmentation.
In this work, we explore a multimodal semi-supervised learning approach for punctuation prediction by learning representations from large amounts of unlabelled audio and text data. Conventional approaches in speech processing typically use forced alignment to encoder per frame acoustic features to word level features and perform multimodal fusion of the resulting acoustic and lexical representations. As an alternative, we explore attention based multimodal fusion and compare its performance with forced alignment based fusion. Experiments conducted on the Fisher corpus show that our proposed approach achieves ~6-9% and ~3-4% absolute improvement (F1 score) over the baseline BLSTM model on reference transcripts and ASR outputs respectively. We further improve the model robustness to ASR errors by performing data augmentation with N-best lists which achieves up to an additional ~2-6% improvement on ASR outputs. We also demonstrate the effectiveness of semi-supervised learning approach by performing ablation study on various sizes of the corpus. When trained on 1 hour of speech and text data, the proposed model achieved ~9-18% absolute improvement over baseline model.