Noise reduction and robustness
Demos and implementations of noise-robust audio and speech signal processing methods.
Online audio noise reduction
This is an experimental, interactive web service for speech and audio noise reduction and enhancement. Users can upload WAV files (10 MB max) and process them using different methods, including novel ones based on noise modulation rate. The WAV files should be PCM-encoded (the usual type of WAV file).
To the online noise reduction system
Acoustic feature post-processing
This is a method to parametrize the typical long-term time dynamics of short-term acoustic parameters and features within some context or signal class. Multiple autoregressive filters on different time scales are trained to represent the typical time behavior. The multi-scale filter can then be used on new data to emphasize class-specific modulation frequencies and to reduce the effect of noise.
References  J. Pohjalainen, P. Alku: Multi-scale modulation filtering in automatic detection of emotions in telephone speech, in Proc. ICASSP, Florence, Italy, May 4-9, 2014. pdf
 J. Pohjalainen, P. Alku: "Filtering and subspace selection for spectral features in detecting speech under physical stress", in Proc. Interspeech, Singapore, September 14-18, 2014. pdf
Robust spectrum analysis
Temporally weighted linear predictive methods have been studied for improving the robustness of speech feature extraction in many applications. Matlab implementations can be found below.
 J. Pohjalainen, P. Alku: "Gaussian mixture linear prediction", in Proc. ICASSP, Florence, Italy, May 4-9, 2014. pdf
 J. Pohjalainen, C. Hanilçi, T. Kinnunen and P. Alku: "Mixture linear prediction in speaker verification under vocal effort mismatch", IEEE Signal Processing Letters, in press. pdf
 J. Pohjalainen, P. Alku: "Extended weighted linear prediction using the autocorrelation snapshot - a robust speech analysis method and its application to recognition of vocal emotions", in Proc. Interspeech, Lyon, France, August 25-29, 2013. pdf
 J. Pohjalainen, R. Saeidi, T. Kinnunen, P. Alku: "Extended weighted linear prediction (XLP) analysis of speech and its application to speaker verification in adverse conditions", in Proc. Interspeech, Makuhari, Japan, September 26-30, 2010. pdf
 C. Magi, J. Pohjalainen, T. Bäckström, P. Alku: ""Stabilised weighted linear prediction"", Speech Communication, 51(5), pp. 401-411, April 2009.
 C. Ma, Y. Kamp, L. F. Willems: "Robust signal selection for linear prediction analysis of voiced speech", Speech Communication 12(2):69--81, 1993.
Matlab code (see also README file) for
Multi-channel audio enhancement
Some examples of smoothing gain functions for musical noise reduction in multi-microphone enhancement applications are shown here.
(by Symeon Delikaris-Manias)
All files are rendered for a monophonic setup using conventional postfiltering methods.
1. Single talker (English) at 0° with different levels of background noise
Last modified: Thu Oct 07 07:54:17 CET 2015