Reliable A posteriori Signal-to-Noise Ratio features selection - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2005

Reliable A posteriori Signal-to-Noise Ratio features selection


This paper adresses the problem of single microphone speech enhancement in noisy environments. State of the art short-time noise reduction techniques are most often expressed as a spectral gain depending on the Signal-to-Noise Ratio (SNR). The well-known decision-directed (DD) approcah drastically limits the level of musical noise but the estimated a priori SNR is biased since it depends on the speech spectrum estimated in the previous frame. The consequence of this biais is an annoying reverberation effect. We propose a new method, called Reliable Features Selection Noise Reduction (RFSNR) technique, that is able to classify the a posteriori SNR estimates into two categories: the reliable features leading to speech components and the unrealiable ones corresponding to musical noise only. Then it is possible to directly enhance speech using a posteriori SNR leading to an unbiased estimator.
Fichier principal
Vignette du fichier
WASPAA-version_finale.pdf (5.04 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

inria-00482648 , version 1 (11-05-2010)


  • HAL Id : inria-00482648 , version 1


Cyril Plapous, Claude Marro, Pascal Scalart. Reliable A posteriori Signal-to-Noise Ratio features selection. IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA), Oct 2005, Mohonc Mountain House, New Paltz, New York, United States. ⟨inria-00482648⟩
297 View
507 Download


Gmail Facebook X LinkedIn More