Title :
Phoneme classification with multinets
Author :
Reynolds, T. Jeff ; Pizzolato, Ednaldo B.
Author_Institution :
Essex Univ., Colchester, UK
Abstract :
The multinet phone classifier architecture is a framework for combining specialised phone detection networks into a posterior probability estimator for all phones. In this paper we give results obtained for the architecture on TIMIT phone classification tasks. We compare it with a standard mixture of Gaussian HMM classifiers
Keywords :
parameter estimation; probability; speech recognition; TIMIT phone classification; automatic speech recognition; multinet phone classifier architecture; multinets; phoneme classification; posterior probability estimator; specialised phone detection networks; Automatic speech recognition; Detectors; Feature extraction; Gaussian processes; Hidden Markov models; Markov processes; Multi-layer neural network; Neural networks; Petroleum; Speech processing;
Conference_Titel :
Signal Processing Proceedings, 1998. ICSP '98. 1998 Fourth International Conference on
Conference_Location :
Beijing
Print_ISBN :
0-7803-4325-5
DOI :
10.1109/ICOSP.1998.770309
