The separation of moving sound sources @ WASPAA & TASLP



Dionyssos Kounades-Bastian, Laurent Girin, Xavier Alameda-Pineda, Sharon Gannot and Radu Horaud


The separation of moving sound sources is a challenging task mainly because it is extremely complex to devise algorithms that robustly discriminate those signal variations due to the intrinsic variation of the sound source from those signal variations due to the time-varying source-to-microphone channel. We successfully investigated this scenario and got the Best Student Paper Award at IEEE WASPAA’15 [1] and an article at IEEE TASLP to appear in 2016 (whose title/abstract read below) [2].

We proposed a novel probabilistic framework based on the complex Gaussian model combined with non-negative matrix factorization (NMF) for sound source separation. The idea is to model the properties associated with moving sources using time-varying mixing filters described by a stochastic temporal process. We present a variational expectation-maximization (VEM) algorithm that employs a Kalman smoother to estimate the mixing filters. The sound sources are separated by means of Wiener filters, built from the estimators provided by the proposed VEM algorithm. Preliminary experiments with simulated data show that, while for static sources we obtain results comparable with the baseline method, in the case of moving source our method outperforms a piece-wise version of the baseline method.

Example: stereo mixture of 4 moving sources.
Original sources (S1, S2, S3, S4), Stereo mixture, Estimated sources (S1, S2, S3, S4).

References

  1. D. Kounades-Bastian, L. Girin, X. Alameda-Pineda, S. Gannot, and R. Horaud, “A Variational EM Algorithm for the Separation of Moving Sound Sources,” in IEEE Workshop on Applications of Signal Processing to Audio and Acoustics, New Paltz, USA, 2015, pp. 1-5. [ bib pdf ] Award Best Student Paper Award
    @inproceedings{Kounades-WASPAA-2015,
      TITLE = {{A Variational EM Algorithm for the Separation of Moving Sound Sources}},
      AUTHOR = {Kounades-Bastian, Dionyssos and Girin, Laurent and Alameda-Pineda, Xavier and Gannot, Sharon and Horaud, Radu},
      BOOKTITLE = {{IEEE Workshop on Applications of Signal Processing to Audio and Acoustics}},
      YEAR = {2015},
      award={Best Student Paper Award},
      pages = {1--5},
      address = {New Paltz, USA},
      pdf={http://xavirema.eu/wp-content/papercite-data/pdf/Kounades-WASPAA-2015.pdf}
    }
  2. D. Kounades-Bastian, L. Girin, X. Alameda-Pineda, S. Gannot, and R. Horaud, “A Variational EM Algorithm for the Separation of Time-Varying Convolutive Audio Mixtures,” IEEE/ACM Transactions on Audio, Speech and Language Processing, vol. 24, iss. 8, pp. 1408-1423, 2016. [ bib pdf arxiv ]
    @article{Kounades-TASLP-2016,
      title={A Variational {EM} Algorithm for the Separation of Time-Varying Convolutive Audio Mixtures},
      author = {Dionyssos Kounades-Bastian and Laurent Girin and Xavier Alameda-Pineda and Sharon Gannot and Radu Horaud},
      journal = {IEEE/ACM Transactions on Audio, Speech and Language Processing},
      year = {2016},
      url = {http://arxiv.org/abs/1510.04595},
      arxiv = {http://arxiv.org/abs/1510.04595},
      volume={24},
      number={8},
      pages={1408-1423},
      pdf={http://xavirema.eu/wp-content/papercite-data/pdf/Kounades-TASLP-2016.pdf}
    }

Category: Research

1 response to: “The separation of moving sound sources @ WASPAA & TASLP“  

  • Taylor says:

    I am often to blogging and i really value your content.
    The article has really peaks my interest. I’m going to bookmark
    your website and keep checking for new tips.

    Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>