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 busy multi-speaker environments.

2 mentions: @wiwer77@Synced_Global
Date: 2021/01/12 22:46

Referring Tweets

@Synced_Global A team from the University of Texas at Austin and Facebook AI Research has introduced VisualVoice, a novel multi-task learning framework #AI #ML #ArtificialIntelligence #MachineLearning t.co/nc4IttX809

Related Entries

Read more Unsupervised Approach for GAN Interpretability Through Semantic Direction Discovery
0 users, 2 mentions 2020/02/12 19:11
Read more MIT's Lincoln Laboratory establishes Biotechnology and Human Systems Division; Spearheads AI and Bio...
0 users, 2 mentions 2020/11/25 23:00
Read more NeurIPS 2020 | Probabilistic Approaches for Algorithmic Recourse With Limited Causal Knowledge | Syn...
0 users, 1 mentions 2020/12/04 17:28
Read more NeurIPS 2020 | Conference Watch on Self-Supervised Learning | Synced
0 users, 2 mentions 2020/12/10 22:12
Read more NeurIPS 2020 Best Papers Released; Google Chief Apologizes for A.I. Researcher’s Dismissal; Hyundai ...
0 users, 1 mentions 2020/12/14 21:51