background and aim : the hearing loss and acoustic feedback signal in the binaural hearing aids may annoy users, make speech reorganization difficult, and affect the sound quality. this paper presents a new method for acoustic feedback suppression in binaural hearing aids, based on the combination of binaural information and blind source separation. this method can enhance sound from the specif...

Our group attracted participants from various research areas to discuss aspects of source separation for audio signals, performed either in a blind way or by using additional knowledge about the underlying sources or the mixing process. In many source separation approaches, one assumes that sources are independent, uncorrelated, and do not overlap with regard to a given representation. Also one...

Municipal solid waste management is a major challenge, especially in metropolises. This research focuses on a non-technical issue in municipal solid waste management named municipal solid waste separation at the source and seeks to find the best policy in terms of model results. Source separation for recycling has been recognized as a way to achieve sustainable municipal solid waste (MSW) manag...

Blind source separation of broad band signals in a multi-path environment remains a di cult problem. Robustness has been limited due to frequency permutation ambiguities. Increasing the number of sensors allows improved performance but introduces degrees of freedom in the separating lters that are not determined by separation criteria. We propose to further shape the lters and improve the robus...

Complexity of music audio signals creates an access problem to specific musical objects or structures within the source samples. Instead of employing more commonly used audio analysis or production techniques to access features, we describe extraction of sub-mixtures from real-world audio using a Probabilistic Latent Component Analysis-based decomposition tool for music composition. This is hig...

In this work, we investigate a method for score-informed source separation using Probabilistic Latent Component Analysis (PLCA). We present extensive test results that give an indication of the performance of the method, its strengths and weaknesses. For this purpose, we created a test database that has been made available to the public, in order to encourage comparisons with alternative methods.

This paper summarizes the audio part of the 2011 communitybased Signal Separation Evaluation Campaign (SiSEC2011). Four speech and music datasets were contributed, including datasets recorded in noisy or dynamic environments and a subset of the SiSEC2010 datasets. The participants addressed one or more tasks out of four source separation tasks, and the results for each task were evaluated using...

We address the problem of blind source separation in the underdetermined mixture case. Two statistical tests are proposed to reduce the number of empirical parameters involved in standard sparsenessbased underdetermined blind source separation (UBSS) methods. The first test performs multisource selection of the suitable time-frequency points for source recovery and is full automatic. The second...

A class of methods based on multichannel linear prediction (MCLP) can achieve effective blind dereverberation of a source, when the source is observed with a microphone array. We propose an inventive use of MCLP as a pre-processing step for blind source separation with a microphone array. We show theoretically that, under certain assumptions, such pre-processing reduces the original blind rever...