Paper availble : Non-linear Prediction using Hierarchical Mixtures of Experts.

[srw1001@eng.cam.ac.uk]

The following paper is available by anonymous ftp from the
archive of the Speech, Vision and Robotics Group at the Cambridge
University Engineering Department and the Neuroprose archives.

			

		NON-LINEAR PREDICTION OF ACOUSTIC VECTORS 
		 USING HIERARCHICAL MIXTURES OF EXPERTS.

		  Steve Waterhouse and Tony Robinson


	    Cambridge University Engineering Department 
		        Trumpington Street 
		        Cambridge CB2 1PZ 
			     England 


                             Abstract

In this paper we consider speech coding as a problem of speech
modelling. In particular, prediction of parameterised speech over
short time segments is performed using the Hierarchical Mixture of
Experts (HME) \cite{JordanJacobs94}. The
HME gives two advantages over traditional non-linear function
approximators such as the Multi-Layer Perceptron (MLP); a statistical
understanding of the operation of the predictor and provision of
information about the performance of the predictor in the form of
likelihood information and local error bars. These two issues are
examined on both toy and real world problems of regression and time
series prediction. In the speech
coding context, we extend the principle of combining local predictions
via the HME to a Vector Quantization scheme in which
fixed local codebooks are combined on-line for each observation.

To appear in Advances in Neural Information Processing Systems 7, edited
by Gerald Tesauro, David Touretzky, and Todd Leen.

************************ How to obtain a copy ************************

a) via ftp from Cambridge University SVR:

unix> ftp svr-ftp.eng.cam.ac.uk
Name: anonymous
Password: (type your email address)
ftp> cd reports
ftp> binary
ftp> get waterhouse_nips94.ps.Z
ftp> quit
unix> uncompress waterhouse_nips94.ps.Z
unix> lpr waterhouse_nips94.ps (or however you print PostScript)

b) via ftp from neuroprose archive:

unix> ftp 
Name: anonymous
Password: (type your email address)
ftp> cd pub/neuroprose/reports
ftp> binary
ftp> get waterhouse.nips94.ps.Z
ftp> quit
unix> uncompress waterhouse.nips94.ps.Z
unix> lpr waterhouse.nips94.ps (or however you print PostScript)


c) or email me: srw1001 at eng.cam.ac.uk

d) (easiest) access my WWW page http://svr-www.eng.cam.ac.uk/~srw1001, 
where the file is symlinked.


-----------------------------------------------------

Steve Waterhouse, Information Engineering,
Cambridge University Engineering Department,
Trumpington Street, Cambridge  CB2 1PZ, UK.
Email: srw1001 at eng.cam.ac.uk  Phone  : (0223) 332800
World Wide Web: http://svr-www.eng.cam.ac.uk/~srw1001