Adaptive Segmentation and Separation of Determined Convolutive Mixtures under Dynamic Conditions

In this paper, we propose a method for blind source separation (BSS) of convolutive audio recordings with short blocks of stationary sources, i.e. dynamically changing source activity but no source movements. It consists of a time-frequency sparseness based localization step to identify segments with stationary sources whose number is equal to the number of microphones. We then use a frequency domain independent component analysis (ICA) algorithm that is robust to short data segments to separate each identified segment. In each segment we solve the permutation problem using the state coherence transform (SCT). Experimental results using real room impulse responses show a good separation performance.

[1]  Bhaskar D. Rao,et al.  Glimpsing independent vector analysis: Separating more sources than sensors using active and inactive states , 2010, 2010 IEEE International Conference on Acoustics, Speech and Signal Processing.

[2]  Rémi Gribonval,et al.  Performance measurement in blind audio source separation , 2006, IEEE Transactions on Audio, Speech, and Language Processing.

[3]  P. Svaizer,et al.  Separating Short Signals in Highly Reverberant Environment by a Recursive Frequency-Domain BSS , 2008, 2008 Hands-Free Speech Communication and Microphone Arrays.

[4]  Juraj Simko,et al.  The CHAINS corpus: CHAracterizing INdividual Speakers , 2006 .

[5]  Dinh-Tuan Pham,et al.  A phase-based dual microphone method to count and locate audio sources in reverberant rooms , 2009, 2009 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics.

[6]  Benedikt Loesch,et al.  Blind Source Separation Based on Time-Frequency Sparseness in the Presence of Spatial Aliasing , 2010, LVA/ICA.

[7]  Satoshi Nakamura,et al.  RWCP Sound Scene Database in Real Acoustic Environment , 2002 .

[8]  Jen-Tzung Chien,et al.  Online Bayesian learning for dynamic source separation , 2010, 2010 IEEE International Conference on Acoustics, Speech and Signal Processing.

[9]  Francesco Nesta,et al.  On the robustness of the multidimensional state coherence transform for solving the permutation problem of frequency-domain ICA , 2010, 2010 IEEE International Conference on Acoustics, Speech and Signal Processing.

[10]  James R. Hopgood,et al.  The effect of sensor placement in blind source separation , 2001, Proceedings of the 2001 IEEE Workshop on the Applications of Signal Processing to Audio and Acoustics (Cat. No.01TH8575).