Audio-Visual Speech Recognition using Deep Learning

Work done during Google Summer of Code

• SyncNet\ : Current speaker recognition in a video frame using Deep Learning, without any labelled speaker data, as described in the paper, Out of time: automated lip sync in the wild (SyncNet)

• LRW_Sentences\ : AVSR model described in the paper, Lip Reading Sentences in the Wild