Reading Group Presentation - SVTS, Scalable Video-to-Speech Synthesis

Short talk where I presented our new scalable video-to-speech model (later published in Interspeech 2022). [recording] [slides]