Speech enhancement and noise PSD estimation

Contact: Richard Hendriks

As speech processing devices like mobile phones, voice controlled devices, and hearing aids have increased in popularity, people expect them to work anywhere and at any time without user intervention. However, the presence of acoustical disturbances limits the use of these applications, degrades their performance, or causes the user difficulties in understanding the conversation or appreciating the device.

A common way to reduce the effects of such disturbances is through the use of single-microphone noise reduction algorithms for speech enhancement.

Items under this topic

Robust joint multi-microphone parameter based on simultaneous confirmatory factor analysis

This software package is meant to estimate jointly all parameters in the multi-microphone setup using confirmatory factor analysis.

Software, Dec 2019

Distributed block-diagonal linearly constrained beamformer

Software, Apr 2019

joint binaural linearly constrained minimum variance (JBLCMV) binaural beamformer

Software, Apr 2019

relaxed binaural linearly constrained minimum variance (RJBLCMV) binaural beamformer

Software, Apr 2019

relaxed binaural linearly constrained minimum variance (RJBLCMV) binaural beamformer using semi-definite convex relaxation (SDCR)

Software, Apr 2019

Algorithm for Noise reduction for speech enhancement

DFT-based single channel noise reduction

Software, Jan 2013

Unbiased MMSE-based Noise Power Estimator

Software, Oct 2011

MMSE-based noise PSD tracker

Software, Mar 2010

GenGam Log MMSE

Toolbox for log-spectral magnitude MMSE estimators of DFT coefficients under super-Gaussian densities

Software, Jan 2009

GenGam

Toolbox for MMSE estimators of DFT coefficients under the generalized Gamma density

Software, Jan 2008