A spatio-temporal neural network applied to visual speech recognition

• Published in: 9th International Conference on Artificial Neural Networks: ICANN '99 pp. 797 - 802
• Main Authors: Baig, A.R, Seguier, R, Vaucher, G
• Format: Conference Proceeding Journal Article
• Language: English
• Published: London IEE 1999
• Subjects: Character recognition; Computer vision and image processing techniques; Learning in AI; Neural computing techniques; Neural nets; Speech processing techniques; Speech recognition; supervised learning; visual speech recognition; character recognition; spatiotemporal neural network; unsupervised learning; speech recognition; Hermitian distance; French digit recognition; feature extraction; automatic lip-reading; learning (artificial intelligence); image recognition; image sequences; neural nets
• ISBN: 0852967217; 9780852967218
• ISSN: 0537-9989
• DOI: 10.1049/cp:19991209

We present a new neural architecture called spatio-temporal neural network (STNN). In this work, we have utilised the Hermitian distance as the basis of spatio-temporal data comparison to adapt a supervised (RCE) and an unsupervised (K-means) learning algorithms for training the STNN weights. A visual speech recognition (automatic lip-reading) system based on STNN is developed and the results obtained on a French digit recognition task are given.