MIR-GAN: Refining Frame-Level Modality-Invariant Representations with Adversarial Network for Audio-Visual Speech Recognition
This work addresses the problem of modality heterogeneity in audio-visual speech recognition, which is an incremental improvement for researchers in multimodal AI.
The paper tackles the challenge of fusing audio and visual modalities in speech recognition by learning shared representations to bridge their distribution gap, and the proposed MIR-GAN method achieves state-of-the-art performance on LRS3 and LRS2 benchmarks.
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.