Title :
A speech recognition method based clustering neural network integration
Author :
Zhang, Jing ; Zhang, Min
Author_Institution :
Dept. of Comput. Sci. & Technol., Guangdong Univ. of Foreign Studies, Guangzhou, China
Abstract :
An improved BP neural network classifier integration method was mainly described, by which using k-means clustering a group of value of weights and thresholds with some differences were gotten, and then as the value of individuals of integrated network to improve the performance of integrated learning, and be successfully applied to non-specific human isolated word speech recognition system. By comparing the experimental result and the traditional Adaboost integration algorithm, the validity of the method was confirmed.
Keywords :
backpropagation; neural nets; pattern clustering; speech recognition; Adaboost integration algorithm; BP neural network classifier integration method; backpropagation neural network; human isolated word speech recognition system; k-means clustering; speech recognition method; Artificial neural networks; Classification algorithms; Clustering algorithms; Predictive models; Speech recognition; Training; Training data; Speech Recognition; difference; integrated learning; k-means clustering;
Conference_Titel :
Electric Information and Control Engineering (ICEICE), 2011 International Conference on
Conference_Location :
Wuhan
Print_ISBN :
978-1-4244-8036-4
DOI :
10.1109/ICEICE.2011.5777537
