End-to-End Multi-Person Audio/Visual Automatic Speech Recognition
This work addresses a realistic bottleneck in A/V ASR for applications like video transcription by enabling multi-person handling, though it is incremental as it builds on existing A/V ASR progress.
The paper tackles the problem of audio-visual automatic speech recognition (A/V ASR) in multi-person scenarios by proposing a fully differentiable model that uses an attention layer to soft-select the appropriate face track, achieving minor word error rate (WER) degradation compared to oracle selection while still benefiting from visual signals.
Traditionally, audio-visual automatic speech recognition has been studied under the assumption that the speaking face on the visual signal is the face matching the audio. However, in a more realistic setting, when multiple faces are potentially on screen one needs to decide which face to feed to the A/V ASR system. The present work takes the recent progress of A/V ASR one step further and considers the scenario where multiple people are simultaneously on screen (multi-person A/V ASR). We propose a fully differentiable A/V ASR model that is able to handle multiple face tracks in a video. Instead of relying on two separate models for speaker face selection and audio-visual ASR on a single face track, we introduce an attention layer to the ASR encoder that is able to soft-select the appropriate face video track. Experiments carried out on an A/V system trained on over 30k hours of YouTube videos illustrate that the proposed approach can automatically select the proper face tracks with minor WER degradation compared to an oracle selection of the speaking face while still showing benefits of employing the visual signal instead of the audio alone.