Title :
Modular neural networks exploit multiple front-ends to improve speech recognition systems
Author :
Antoniou, Christos A. ; Reynolds, T. Jeff
Author_Institution :
Dept. of Comput. Sci., Essex Univ., Colchester, UK
Abstract :
We have been investigating the possible advantages of a modular/ensemble neural network for acoustic modelling. We report experiments with ensembles of networks trained on data provided by different front-end preprocessing methods. As for previous work we train a network ensemble for each individual phone and combine the outputs of the ensemble using a further trained network. The combined system provides significant improvements for phone recognition and classification on the TIMIT corpus. Our results are now better than the best context-independent systems in the literature and close to the best context-dependent systems
Keywords :
acoustic signal processing; neural nets; signal classification; speech recognition; TIMIT corpus; acoustic modelling; context-dependent systems; context-independent systems; ensemble neural network; front-end preprocessing methods; modular neural network; multiple front-ends; phone classification; phone recognition; speech recognition systems; Acoustic scattering; Computer science; Feature extraction; Hidden Markov models; History; Intelligent systems; Neural networks; Speech processing; Speech recognition; Training data;
Conference_Titel :
Knowledge-Based Intelligent Engineering Systems and Allied Technologies, 2000. Proceedings. Fourth International Conference on
Conference_Location :
Brighton
Print_ISBN :
0-7803-6400-7
DOI :
10.1109/KES.2000.885793