Audio-visual speech recognition (AVSR) attracts a surge of research interest recently by leveraging multimodal signals to understand human speech. Mainstream approaches addressing this task have developed sophisticated architectures and techniques for multi-modality fusion and representation learning. However, the natural heterogeneity of different modalities causes distribution gap between their representations, making it challenging to fuse them. In this paper, we aim to learn the shared representations across modalities to bridge their gap. Different from existing similar methods on other multimodal tasks like sentiment analysis, we focus on the temporal contextual dependencies considering the sequence-to-sequence task setting of AVSR. In particular, we propose an adversarial network to refine frame-level modality-invariant representations (MIR-GAN), which captures the commonality across modalities to ease the subsequent multimodal fusion process. Extensive experiments on public benchmarks LRS3 and LRS2 show that our approach outperforms the state-of-the-arts.

翻译：视听语音识别（AVSR）通过利用多模态信号理解人类语音，近年来引发了研究热潮。主流方法针对该任务开发了复杂的架构与技术以实现多模态融合与表示学习。然而，不同模态的天然异质性导致其表示存在分布差异，使得融合面临挑战。本文旨在学习跨模态的共享表示以弥合这一差异。与情感分析等其他多模态任务中现有方法不同，我们着眼于AVSR序列到序列任务设置下的时间上下文依赖关系。具体而言，我们提出一种对抗网络来优化帧级模态不变表示（MIR-GAN），通过捕获模态间的共性以简化后续多模态融合过程。在公开基准数据集LRS3和LRS2上的大量实验表明，本方法性能优于现有最先进技术。