Perform research to advance the state-of-the-art in Video ASR (automatic speech recognition). Investigate self-supervised learning (SSL) to create audio representations that are indicative of how users will react, based on their comments and hashtags. Explore applications of SSL to hate-speech detection, the detection of suicide and self-injury, and other integrity applications. Guide full-time research scientists and software engineers to develop multimodal/SSL techniques in the context of Facebook's unique data assets.
PhD in Computer Science or related field