Neural Networks

Image Analysis and Processing

Volume 974 of the series Lecture Notes in Computer Science pp 152-158


A visual speech model based on fuzzy-neuro methods

  • Hans H. BotheAffiliated withDepartment of Electronics, Technical University of Berlin

* Final gross prices may vary according to local VAT.

Get Access


This paper describes a new approach of modeling visual speech movements, based on a codebook of characteristic key-pictures and a complex fuzzy neural network (FNN). Goal is the development of a computer animation program as a training aid for learning lip-reading. The network architecture makes possible a fusion of linguistic expert knowledge into the FNN. The current PC version allows a synchronization of the animation program with a special stand-alone speech synthesis computer via a Centronics parallel interface.