I am researching with the University of Newcastle upon Tyne in the field of Neural
Network Signal Processing. The intention is to utilise the nets to control a computer
generated image of a human face so as to provide lip synchronisation with a vocal
sound-track. The research is concerned with the architecture and algorithmic
processing of the net and I do not wish to develop the input and output features of the
system. Therefore there exist a few items which I would rather simple acquire and use
instead of spending time creating them. If anyone could assist me in these matters it
would be greatly appreciated.
1. Phoneme recogniser from textual and/or vocal input
2. Wire frame model of the human face (a polygonal structure representing the
face in which the user can move certain 'feature' points in order to create different
facial and lip sync. images)
In addition I would be interested to hear from anyone with knowledge in the field of:
1. Grammar recognition and sentence structure segmentation
2. Speech recognition with a view to segment sentences and extract phoneme
content
3. Image compression and recognition with particular reference to the human
face
4. Neural network packages available commercially or from the Web
I will be presenting some of my work at the University of Cambridge Programme for
Industry Neural Network Summer School 24 to 27 July 1995 if anyone is interested,
otherwise I am current constructing a Web page and I can provide you with the URL.
Thank you for taking the time to read this message
The message has been replicated in other newsgroups and I am sorry if you
inadvertently come across it again.
Christian
C.M.Jones at newcastle.ac.uk