Open AccessMultistream neural architectures for cued-speech recognition using a pre-trained visual feature extractor and constrained CTC decoding
Author(s) -
Sanjana Sankar,
Denis Beautemps,
Thomas Hueber
Publication year - 2022
Publication title -
hal (le centre pour la communication scientifique directe)
Language(s) - English
Resource type - Conference proceedings
Subject(s) - computer science , speech recognition , decoding methods , extractor , feature (linguistics) , artificial intelligence , feature extraction , pattern recognition (psychology) , cued speech , artificial neural network , engineering , psychology , telecommunications , linguistics , philosophy , process engineering , cognitive psychology