SPE-85.6
MULTISTREAM NEURAL ARCHITECTURES FOR CUED SPEECH RECOGNITION USING A PRE-TRAINED VISUAL FEATURE EXTRACTOR AND CONSTRAINED CTC DECODING
Sanjana Sankar, Denis Beautemps, Thomas Hueber, Centre National de la Recherche Scientifique, France
Session:
Sign Language and Lip Reading
Track:
Speech and Language Processing
Location:
Gather Area E
Presentation Time:
Fri, 13 May, 21:00 - 21:45 China Time (UTC +8)
Fri, 13 May, 13:00 - 13:45 UTC
Fri, 13 May, 13:00 - 13:45 UTC
Session Chair:
Eric Fosler-Lussier, The Ohio State University
Session SPE-85
SPE-85.1: PHONOLOGY RECOGNITION IN AMERICAN SIGN LANGUAGE
Federico Tavella, Aphrodite Galata, Angelo Cangelosi, The University of Manchester, United Kingdom of Great Britain and Northern Ireland
SPE-85.2: SPATIO-TEMPORAL GRAPH CONVOLUTIONAL NETWORKS FOR CONTINUOUS SIGN LANGUAGE RECOGNITION
Maria Parelli, Petros Maragos, National Technical University of Athens, Greece; Katerina Papadimitriou, Gerasimos Potamianos, University of Thessaly, Greece; Georgios Pavlakos, University of California, United States of America
SPE-85.3: SENSORS TO SIGN LANGUAGE: A NATURAL APPROACH TO EQUITABLE COMMUNICATION
Thomas Fouts, University of Michigan, United States of America; Ali Hindy, Stanford University, United States of America; Chris Tanner, Harvard University, United States of America
SPE-85.4: ACCURATE AND RESOURCE-EFFICIENT LIPREADING WITH EFFICIENTNETV2 AND TRANSFORMERS
Alexandros Koumparoulis, Gerasimos Potamianos, University of Thessaly, Greece
SPE-85.5: Training Strategies For Improved Lip-reading
Pingchuan Ma, Yujiang Wang, Stavros Petridis, Jie Shen, Maja Pantic, Imperial College London, United Kingdom of Great Britain and Northern Ireland
SPE-85.6: MULTISTREAM NEURAL ARCHITECTURES FOR CUED SPEECH RECOGNITION USING A PRE-TRAINED VISUAL FEATURE EXTRACTOR AND CONSTRAINED CTC DECODING
Sanjana Sankar, Denis Beautemps, Thomas Hueber, Centre National de la Recherche Scientifique, France