CONF lathoud05c/IDIAP A Spectrogram Model for Enhanced Source Localization and Noise-Robust ASR Lathoud, Guillaume Magimai.-Doss, Mathew Mesot, Bertrand EXTERNAL https://publications.idiap.ch/attachments/papers/2005/lathoud05c.pdf PUBLIC https://publications.idiap.ch/index.php/publications/showcite/lathoud-rr-05-13 Related documents Proceedings of INTERSPEECH 2005 2005 Lisbon, Portugal September 2005 IDIAP-RR 05-13 This paper proposes a simple, computationally efficient 2-mixture model approach to discrimination between speech and background noise. It is directly derived from observations on real data, and can be used in a fully unsupervised manner, with the EM algorithm. A first application to sector-based, joint audio source localization and detection, using multiple microphones, confirms that the model can provide major enhancement. A second application to the single channel speech recognition task in a noisy environment yields major improvement on stationary noise and promising results on non-stationary noise. REPORT lathoud-rr-05-13/IDIAP A Frequency-Domain Silence Noise Model Lathoud, Guillaume Magimai.-Doss, Mathew Mesot, Bertrand EXTERNAL https://publications.idiap.ch/attachments/reports/2005/rr-05-13.pdf PUBLIC Idiap-RR-13-2005 2005 IDIAP Martigny, Switzerland To appear in ``Proceedings of INTERSPEECH 2005'' This paper proposes a simple, computationally efficient 2-mixture model approach to discrimination between speech and background noise. It is directly derived from observations on real data, and can be used in a fully unsupervised manner, with the EM algorithm. A first application to sector-based, joint audio source localization and detection, using multiple microphones, confirms that the model can provide major enhancement. A second application to the single channel speech recognition task in a noisy environment yields major improvement on stationary noise and promising results on non-stationary noise.