Title :
State based sub-band LP Wiener filters for speech enhancement in car environments
Author :
Chen, Aimin ; Vaseghi, Saeed ; McCourt, Paul
Author_Institution :
Queen´´s Univ., Belfast, UK
Abstract :
The performance of Wiener filters in restoring the quality and intelligibility of noisy speech depends on: (i) the accuracy of the estimates of the power spectra or the correlation values of the noise and the speech processes, and (ii) on the Wiener filter structure. In this paper a Bayesian method is proposed where model combination and model decomposition are employed for the estimation of parameters required to implement subband LP Wiener filters. The use of subband LP Wiener filters provides advantages in terms of improved parameter estimates and also in restoring the temporal-spectral composition of speech. The method is evaluated, and compared with the parallel model combination, using the TIMIT continuous speech database with BMW and VOLVO car noise databases
Keywords :
Bayes methods; Wiener filters; acoustic noise; cascade networks; correlation methods; hidden Markov models; land mobile radio; parameter estimation; prediction theory; spectral analysis; speech enhancement; speech intelligibility; speech recognition; Bayesian method; car environments; car noise; correlation values; intelligibility; noisy speech; parallel model combination; parameter estimates; power spectra; quality; speech enhancement; state based sub-band LP Wiener filters; temporal-spectral composition; Databases; Hidden Markov models; Parameter estimation; Predictive models; Signal processing; Speech enhancement; Speech processing; Speech recognition; Wiener filter; Working environment noise;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 2000. ICASSP '00. Proceedings. 2000 IEEE International Conference on
Conference_Location :
Istanbul
Print_ISBN :
0-7803-6293-4
DOI :
10.1109/ICASSP.2000.861919
