Cross-Modal Mutual Learning for Cued Speech Recognition
This work addresses visual communication for hearing-impaired people by improving Cued Speech Recognition, introducing a novel method for multi-modal interaction and the first large-scale dataset for Mandarin Chinese.
The paper tackles the problem of asynchronous modalities in Automatic Cued Speech Recognition by proposing a transformer-based cross-modal mutual learning framework, which achieves superior recognition performance over state-of-the-art methods across multiple languages.
Automatic Cued Speech Recognition (ACSR) provides an intelligent human-machine interface for visual communications, where the Cued Speech (CS) system utilizes lip movements and hand gestures to code spoken language for hearing-impaired people. Previous ACSR approaches often utilize direct feature concatenation as the main fusion paradigm. However, the asynchronous modalities i.e., lip, hand shape and hand position) in CS may cause interference for feature concatenation. To address this challenge, we propose a transformer based cross-modal mutual learning framework to prompt multi-modal interaction. Compared with the vanilla self-attention, our model forces modality-specific information of different modalities to pass through a modality-invariant codebook, collating linguistic representations for tokens of each modality. Then the shared linguistic knowledge is used to re-synchronize multi-modal sequences. Moreover, we establish a novel large-scale multi-speaker CS dataset for Mandarin Chinese. To our knowledge, this is the first work on ACSR for Mandarin Chinese. Extensive experiments are conducted for different languages i.e., Chinese, French, and British English). Results demonstrate that our model exhibits superior recognition performance to the state-of-the-art by a large margin.