Model-driven detection of Clean Speech Patches in Noise

Jonathan Laidler, Martin CookeNeil D. Lawrence
Proceedings of Interspeech 2007, 2007.

Abstract

Listeners may be able to recognise speech in adverse conditions by glimpsing time-frequency regions where the target speech is dominant. Previous computational attempts to identify such regions have been source-driven, using primitive cues. This paper describes a model-driven approach in which the likelihood of spectro-temporal patches of a noisy mixture representing speech is given by a generative model. The focus is on patch size and patch modelling. Small patches lead to a lack of discrimination, while large patches are more likely to contain contributions from other sources. A cleanness measure reveals that a good patch size is one which extends over a quarter of the speech frequency range and lasts for 40 ms. Gaussian mixture models are used to represent patches. A compact representation based on a 2D discrete cosine transform leads to reasonable speech/background discrimination.

Cite this Paper


BibTeX
@InProceedings{Laidler:model07, title = {Model-driven detection of Clean Speech Patches in Noise}, author = {Laidler, Jonathan and Cooke, Martin and Lawrence, Neil D.}, booktitle = {Proceedings of Interspeech 2007}, year = {2007}, url = {http://inverseprobability.com/publications/laidler-model07.html}, abstract = {Listeners may be able to recognise speech in adverse conditions by glimpsing time-frequency regions where the target speech is dominant. Previous computational attempts to identify such regions have been source-driven, using primitive cues. This paper describes a model-driven approach in which the likelihood of spectro-temporal patches of a noisy mixture representing speech is given by a generative model. The focus is on patch size and patch modelling. Small patches lead to a lack of discrimination, while large patches are more likely to contain contributions from other sources. A cleanness measure reveals that a good patch size is one which extends over a quarter of the speech frequency range and lasts for 40 ms. Gaussian mixture models are used to represent patches. A compact representation based on a 2D discrete cosine transform leads to reasonable speech/background discrimination.}, note = {} }
Endnote
%0 Conference Paper %T Model-driven detection of Clean Speech Patches in Noise %A Jonathan Laidler %A Martin Cooke %A Neil D. Lawrence %B Proceedings of Interspeech 2007 %D 2007 %F Laidler:model07 %U http://inverseprobability.com/publications/laidler-model07.html %X Listeners may be able to recognise speech in adverse conditions by glimpsing time-frequency regions where the target speech is dominant. Previous computational attempts to identify such regions have been source-driven, using primitive cues. This paper describes a model-driven approach in which the likelihood of spectro-temporal patches of a noisy mixture representing speech is given by a generative model. The focus is on patch size and patch modelling. Small patches lead to a lack of discrimination, while large patches are more likely to contain contributions from other sources. A cleanness measure reveals that a good patch size is one which extends over a quarter of the speech frequency range and lasts for 40 ms. Gaussian mixture models are used to represent patches. A compact representation based on a 2D discrete cosine transform leads to reasonable speech/background discrimination. %Z
RIS
TY - CPAPER TI - Model-driven detection of Clean Speech Patches in Noise AU - Jonathan Laidler AU - Martin Cooke AU - Neil D. Lawrence BT - Proceedings of Interspeech 2007 DA - 2007/08/01 ID - Laidler:model07 UR - http://inverseprobability.com/publications/laidler-model07.html AB - Listeners may be able to recognise speech in adverse conditions by glimpsing time-frequency regions where the target speech is dominant. Previous computational attempts to identify such regions have been source-driven, using primitive cues. This paper describes a model-driven approach in which the likelihood of spectro-temporal patches of a noisy mixture representing speech is given by a generative model. The focus is on patch size and patch modelling. Small patches lead to a lack of discrimination, while large patches are more likely to contain contributions from other sources. A cleanness measure reveals that a good patch size is one which extends over a quarter of the speech frequency range and lasts for 40 ms. Gaussian mixture models are used to represent patches. A compact representation based on a 2D discrete cosine transform leads to reasonable speech/background discrimination. N1 - ER -
APA
Laidler, J., Cooke, M. & Lawrence, N.D.. (2007). Model-driven detection of Clean Speech Patches in Noise. Proceedings of Interspeech 2007 Available from http://inverseprobability.com/publications/laidler-model07.html.

Related Material