%Aigaion2 BibTeX export from Idiap Publications %Thursday 21 November 2024 04:56:04 PM @INPROCEEDINGS{lathoud05e, author = {Lathoud, Guillaume and Magimai.-Doss, Mathew and Mesot, Bertrand and Bourlard, Herv{\'{e}}}, projects = {Idiap}, month = {12}, title = {Unsupervised {S}pectral {S}ubtraction for Noise-{R}obust {ASR}}, booktitle = {Proceedings of the 2005 {IEEE} {ASRU} {W}orkshop}, year = {2005}, address = {San Juan, Puerto Rico}, note = {IDIAP RR 05-42}, crossref = {lathoud-rr-05-42}, abstract = {This paper proposes a simple, computationally efficient 2-mixture model approach to discriminate between speech and background noise at the magnitude spectrogram level. It is directly derived from observations on real data, and can be used in a fully unsupervised manner, with the EM algorithm. In this paper, the 2-mixture model is used in an ``Unsupervised Spectral Subtraction'' scheme that can be applied as a pre-processing step for any acoustic feature extraction scheme, such as MFCCs or PLP. The goal is to improve noise-robustness of the acoustic features. Experimental results on both OGI Numbers 95 and Aurora 2 tasks yielded a major improvement on all noise conditions, while retaining a similar performance on clean conditions.}, pdf = {https://publications.idiap.ch/attachments/papers/2005/lathoud05e.pdf}, postscript = {ftp://ftp.idiap.ch/pub/papers/2005/lathoud05e.ps.gz}, ipdinar={2005}, ipdmembership={speech}, } crossreferenced publications: @TECHREPORT{lathoud-rr-05-42, author = {Lathoud, Guillaume and Magimai.-Doss, Mathew and Mesot, Bertrand and Bourlard, Herv{\'{e}}}, projects = {Idiap}, title = {Unsupervised {S}pectral {S}ubstraction for Noise-{R}obust {ASR}}, type = {Idiap-RR}, number = {Idiap-RR-42-2005}, year = {2005}, institution = {IDIAP}, address = {Martigny, Switzerland}, note = {Published in Proceedings of the 2005 IEEE ASRU Workshop}, abstract = {This paper proposes a simple, computationally efficient \mbox{2-mixture} model approach to discriminate between speech and background noise at the magnitude spectrogram level. It is directly derived from observations on real data, and can be used in a fully unsupervised manner, with the EM algorithm. In this paper, the 2-mixture model is used in an ``Unsupervised Spectral Substraction'' scheme that can be applied as a pre-processing step for any acoustic feature extraction scheme, such as MFCCs or PLP. The goal is to improve noise-robustness of the acoustic features. Experimental results on both OGI~Numbers~95 and Aurora~2 tasks yielded a major improvement on all noise conditions, while retaining a similar performance on clean conditions.}, pdf = {https://publications.idiap.ch/attachments/reports/2005/rr-05-42.pdf}, postscript = {ftp://ftp.idiap.ch/pub/reports/2005/rr-05-42.ps.gz}, ipdinar={2005}, ipdmembership={speech}, language={English}, }