VisualVoice Uses Facial Appearance to Boost SOTA in Speech Separation | Synced
Recent AI research on speech separation has explored ways to associate lip motions in videos with audio, but this approach suffers when speakers’ lips are occluded, which they often are in bu...
- computer vision & graphics
- machine learning & data science
- research
- computer vision
- facebook ai research
Source: Synced | AI Technology & Industry Review
Recent AI research on speech separation has explored ways to associate lip motions in videos with audio, but this approach suffers when speakers’ lips are occluded, which they often are in busy multi-speaker environments.