Title :
Research on a kind of Noisy Tibetan speech recognition algorithm based on WNN
Author :
Yong Lu ; Haining Huang
Author_Institution :
14th Res. Lab., Chinese Acad. Sci., Beijing, China
Abstract :
The research on noisy Tibetan speech recognition algorithm based on wavelet neural network (WNN) combined with auditory feature was carried out in this paper. The recognition classifier based on WNN was designed, and Mel Frequency Cepstrum Constant (MFCC) feature was given. Then the simulation on the given algorithm was run under the different signal to noise ratios (SNR), and the results illustrated that the presented algorithm is effect.
Keywords :
feedforward neural nets; pattern classification; speech recognition; wavelet transforms; auditory feature; mel frequency cepstrum constant feature; noisy Tibetan speech recognition algorithm; recognition classifier; signal to noise ratios; wavelet neural network; Feature extraction; Hidden Markov models; Mathematical model; Mel frequency cepstral coefficient; Signal to noise ratio; Speech; Speech recognition; Tibetan speech processing; neural network; recognition; wavelet;
Conference_Titel :
Natural Computation (ICNC), 2011 Seventh International Conference on
Conference_Location :
Shanghai
Print_ISBN :
978-1-4244-9950-2
DOI :
10.1109/ICNC.2011.6022224
