ViCocktail: Automated Multi-Modal Data Collection for Vietnamese Audio-Visual Speech Recognition
This work addresses the problem of limited AVSR data for under-resourced languages, offering an incremental improvement in dataset collection efficiency.
The authors tackled the scarcity of datasets for Audio-Visual Speech Recognition (AVSR) in non-English languages by developing an automated method to generate AVSR datasets from raw video, and demonstrated its effectiveness by creating a baseline model for Vietnamese that achieves competitive performance in clean conditions and significantly outperforms audio-only systems in noisy environments.
Audio-Visual Speech Recognition (AVSR) has gained significant attention recently due to its robustness against noise, which often challenges conventional speech recognition systems that rely solely on audio features. Despite this advantage, AVSR models remain limited by the scarcity of extensive datasets, especially for most languages beyond English. Automated data collection offers a promising solution. This work presents a practical approach to generate AVSR datasets from raw video, refining existing techniques for improved efficiency and accessibility. We demonstrate its broad applicability by developing a baseline AVSR model for Vietnamese. Experiments show the automatically collected dataset enables a strong baseline, achieving competitive performance with robust ASR in clean conditions and significantly outperforming them in noisy environments like cocktail parties. This efficient method provides a pathway to expand AVSR to more languages, particularly under-resourced ones.