Login / Signup

Audio-Visual Speech Enhancement Method Conditioned in the Lip Motion and Speaker-Discriminative Embeddings.

Koichiro ItoMasaaki YamamotoKenji Nagamatsu
Published in: ICASSP (2021)
Keyphrases
  • audio visual
  • information retrieval
  • multi modal
  • visual data
  • audio visual speech recognition
  • feature extraction
  • probabilistic model
  • visual features
  • high dimensional data
  • speaker verification