 |
International Workshop on Hands-Free Speech Communication (HSC2001)April 9-11, 2001 |
 |
|
International Workshop on Hands-Free Speech Communication (HSC2001)April 9-11, 2001 |
|
n this paper, we introduce a Minima Controlled Recursive Averaging (MCRA) noise estimation approach for robust speech enhancement. The noise spectrum is estimated by recursively averaging past spectral power values, using a smoothing parameter that is adjusted by the signal presence probability in subbands We show that presence of speech in a given frame of a subband can be determined by the ratio between the local energy of the noisy speech and its minimum within a specified time window. The noise estimate is unbiased, computationally efficient, robust with respect to the input signal-to-noise ratio and type of underlying additive noise, and characterized by the ability to quickly follow abrupt changes in the noise spectrum. Incorporated in the Optimally-Modified Log-Spectral Amplitude estimator, excellent noise suppression is achieved, while retaining week speech components and avoiding the musical residual noise phenomena.
Bibliographic reference. Cohen, Israel / Berdugo, Baruch (2001): "Spectral enhancement by tracking speech presence probability in subbands", In HSC2001, 95-98.