Title :
The perceptual wavelet feature for noise robust Vietnamese speech recognition
Author :
Quoc Trung, Nguyen ; Trung Nghia, Phung
Author_Institution :
Fac. of Electron. & Telecommun., Hanoi Univ. of Technol., Hanoi
Abstract :
This study proposes a novel noise robust speech feature for Vietnamese speech recognition based on the Bark scale, the Perceptual Wavelet Packet Transform (PWPT), the Wavelet sub-band Energy Parameters (WEP). This method is used in compare to the widely used MFCC feature recognizer. The experimental results show that our proposal is superior to the widely used MFCC feature for Vietnamese speech corpus with both clean and white noisy speech.
Keywords :
speech recognition; wavelet transforms; Bark scale; PWPT; Vietnamese speech recognition; WEP; perceptual wavelet feature; perceptual wavelet packet transform; wavelet subband energy parameters; Discrete cosine transforms; Fourier transforms; Mel frequency cepstral coefficient; Noise robustness; Speech enhancement; Speech processing; Speech recognition; Wavelet packets; Wavelet transforms; Working environment noise; Noise Robust Speech Recognition; Wavelet;
Conference_Titel :
Communications and Electronics, 2008. ICCE 2008. Second International Conference on
Conference_Location :
Hoi an
Print_ISBN :
978-1-4244-2425-2
Electronic_ISBN :
978-1-4244-2426-9
DOI :
10.1109/CCE.2008.4578968
