Title :
On minimum mean square error speech enhancement
Author_Institution :
AT&T Bell Lab., Murray Hill, NJ, USA
Abstract :
Motivation for using minimum mean square error (MMSE) estimation in noisy speech enhancement problems is given. An MMSE estimator which is based on hidden Markov modeling of the clean signal as well as the noise process is systematically developed. The MMSE estimator is tested and compared with the spectral subtraction estimator in vector quantization of noisy speech signals
Keywords :
Markov processes; data compression; encoding; noise; parameter estimation; speech analysis and processing; speech intelligibility; MMSE estimator; hidden Markov modeling; minimum mean square error; noisy speech enhancement; spectral subtraction estimator; vector quantization; Distortion measurement; Estimation error; Gaussian noise; Hidden Markov models; Mean square error methods; Noise measurement; Parameter estimation; Particle measurements; Pollution measurement; Speech enhancement;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1991. ICASSP-91., 1991 International Conference on
Conference_Location :
Toronto, Ont.
Print_ISBN :
0-7803-0003-3
DOI :
10.1109/ICASSP.1991.150509
