DocumentCode :
2215410
Title :
Kalman filter with linear predictor and harmonic noise models for noisy speech enhancement
Author :
Qin Yan ; Vaseghi, Saeed ; Zavarehei, Esfandiar ; Milner, Ben
Author_Institution :
Sch. of Eng. & Design, Brunel Univ., Uxbridge, UK
fYear :
2006
fDate :
4-8 Sept. 2006
Firstpage :
1
Lastpage :
4
Abstract :
This paper presents a method for noisy speech enhancement based on integration of a formant-tracking linear prediction (FTLP) model of spectral envelope and a harmonic noise model (HNM) of the excitation of speech. The time-varying trajectories of the parameters of the LP and HNM models are tracked with Viterbi classifiers and smoothed with Kalman filters. A frequency domain pitch estimation is proposed, that searches for the peak SNRs at the harmonics. The LP-HNM model is used to deconstruct noisy speech, de-noise its LP and HNM models and then reconstitute cleaned speech. Experimental evaluations show the performance gains resulting from the formant tracking, harmonic extraction and noise reduction stages.
Keywords :
Kalman filters; linear predictive coding; speech enhancement; Kalman filter; Viterbi classifiers; formant-tracking linear prediction model; frequency domain pitch estimation; harmonic extraction; harmonic noise models; linear predictor; noise reduction stages; noisy speech deconstruction; noisy speech enhancement; spectral envelope; time-varying trajectories; Europe; Harmonic analysis; Hidden Markov models; Kalman filters; Noise measurement; Speech; Speech enhancement;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Signal Processing Conference, 2006 14th European
Conference_Location :
Florence
ISSN :
2219-5491
Type :
conf
Filename :
7071212
Link To Document :
بازگشت