Leveraging Unimodal Self-Supervised Learning for Multimodal Audio-Visual Speech Recognition
Training Transformer-based models demands a large amount of data, while obtaining aligned and labelled data in multimodality is rather cost-demanding, especially for audio-visual speech recognition (AVSR). Thus it makes a lot of sense to make use of unlabelled unimodal data. On the other side, although the effectiveness of large-scale self-supervised learning is well established in both audio and visual modalities, how to integrate those pre-trained models into a multimodal scenario remains underexplored. In this work, we successfully leverage unimodal self-supervised learning to promote the multimodal AVSR. In particular, audio and visual front-ends are trained on large-scale unimodal datasets, then we integrate components of both front-ends into a larger multimodal framework which learns to recognize parallel audio-visual data into characters through a combination of CTC and seq2seq decoding. We show that both components inherited from unimodal self-supervised learning cooperate well, resulting in that the multimodal framework yields competitive results through fine-tuning. Our model is experimentally validated on both word-level and sentence-level tasks. Especially, even without an external language model, our proposed model raises the state-of-the-art performances on the widely accepted Lip Reading Sentences 2 (LRS2) dataset by a large margin, with a relative improvement of 30%.
PDF Abstract ACL 2022 PDF ACL 2022 Abstract
Datasets
ImageNet
LibriSpeech
LRW
LRS2
| Task | Dataset | Model | Metric Name | Metric Value | Global Rank | Benchmark |
|---|---|---|---|---|---|---|
| Lipreading | Lip Reading in the Wild | MoCo + Wav2Vec by SJTU LUMIA | Top-1 Accuracy | 85.0 | # 9 | |
| Audio-Visual Speech Recognition | LRS2 | MoCo + wav2vec by SJTU LUMIA (w/o extLM) | Test WER | 2.6 | # 1 | |
| Lipreading | LRS2 | MoCo + wav2vec by SJTU LUMIA (w/o extLM) | Word Error Rate (WER) | 43.2% | # 6 | |
| Automatic Speech Recognition | LRS2 | MoCo + wav2vec by SJTU LUMIA (w/o extLM) | Test WER | 2.7 | # 1 |
