Improving Audiovisual Speech Recognition through Synthetic Visual Data Augmentation
This work explores the use of synthetic visual data as a solution, using an audio-driven talking-head pipeline to generate lip-synchronized visual content from existing audio data and demonstrates that synthetic data alone can serve as a baseline for AVSR training in languages lacking AV datasets.
Pol Buitrago, Pol Gàlvez, Javier Hernando
· 1 citation
· ⚡1