Title :
Biomimetic Pattern Recognition for Speaker-Independent Speech Recognition
Author :
Qin, Hong ; Wang, Shoujue ; Sun, Hua
Author_Institution :
Inst. of Semicond., Chinese Acad. of Sci., Beijing
Abstract :
In speaker-independent speech recognition, the disadvantage of the most diffused technology (hidden Markov models) is not only the need of many more training samples, but also long train time requirement. This paper describes the use of biomimetic pattern recognition (BPR) in recognizing some Mandarin speech in a speaker-independent manner. The vocabulary of the system consists of 15 Chinese dish´s names. Neural networks based on multi-weight neuron (MWN) model are used to train and recognize the speech sounds. Experimental results are presented to show that the system, which can carry out real time recognition of the persons from different provinces speaking common Chinese speech, outperforms HMMs especially in the cases of samples of a finite size
Keywords :
biomimetics; hidden Markov models; natural languages; neural nets; speaker recognition; Mandarin speech; biomimetic pattern recognition; hidden Markov models; multiweight neuron model; neural networks; speaker-independent speech recognition; Automatic speech recognition; Biomimetics; Business process re-engineering; Hidden Markov models; Neural networks; Neurons; Pattern recognition; Real time systems; Speech recognition; Vocabulary; Biomimetic Pattern Recognition; Dynamic Time Warping; Hidden Markov Models; Speech Recognition;
Conference_Titel :
Neural Networks and Brain, 2005. ICNN&B '05. International Conference on
Conference_Location :
Beijing
Print_ISBN :
0-7803-9422-4
DOI :
10.1109/ICNNB.2005.1614846
