Audio-Visual Speech Recognition
Audio-Visual Speech Recognition — Combines acoustic speech with lip and facial motion to improve speech-recognition robustness.
Technical Context
Combines acoustic speech with lip and facial motion to improve speech-recognition robustness.
Temporal alignment between audio and video is fundamental; inter-modal delay can propagate directly as error into the fusion stage.