arXiv - CSCL: "MixSpeech: Cross-Modality Self-Learning with Audi…" - Qoto Mastodon

arXiv - CSCL @arxiv_cscl@qoto.org

MixSpeech: Cross-Modality Self-Learning with Audio-Visual Stream Mixup for Visual Speech Translation and Recognition. (arXiv:2303.05309v1 [cs.CV])

http://arxiv.org/abs/2303.05309 #arXiv #NLProc

Mar 10, 2023, 03:15 · · arxiv-cscl · · ·

Sign in to participate in the conversation