A latently constrained mixture model for audio source separation and localization
Deleforge A, Horaud R (2012)
Publication Language: English
Publication Type: Conference contribution
Publication year: 2012
Publisher: Springer
City/Town: Berlin, Heidelberg
Pages Range: 372-379
Conference Proceedings Title: Proceedings of th 10th International Conference on Latent Variable Analysis and Source Separation (LVA/ICA), 2012
Event location: Tel Aviv
ISBN: 978-3-642-28551-6
DOI: 10.1007/978-3-642-28551-6_46
Abstract
We present a method for audio source separation and localization from binaural recordings. The method combines a new generative probabilistic model with time-frequency masking. We suggest that device-dependent relationships between point-source positions and interaural spectral cues may be learnt in order to constrain a mixture model. This allows to capture subtle separation and localization features embedded in the auditory data. We illustrate our method with data composed of two and three mixed speech signals in the presence of reverberations. Using standard evaluation metrics, we compare our method with a recent binaural-based source separation-localization algorithm.
Authors with CRIS profile
How to cite
APA:
Deleforge, A., & Horaud, R. (2012). A latently constrained mixture model for audio source separation and localization. In Proceedings of th 10th International Conference on Latent Variable Analysis and Source Separation (LVA/ICA), 2012 (pp. 372-379). Tel Aviv, IL: Berlin, Heidelberg: Springer.
MLA:
Deleforge, Antoine, and Radu Horaud. "A latently constrained mixture model for audio source separation and localization." Proceedings of the 10th International Conference on Latent Variable Analysis and Signal Separation, Tel Aviv Berlin, Heidelberg: Springer, 2012. 372-379.
BibTeX: Download