Unsupervised Spectral Subtraction for Noise-Robust ASR
| Type of publication: | Conference paper |
| Citation: | lathoud05e |
| Booktitle: | Proceedings of the 2005 IEEE ASRU Workshop |
| Year: | 2005 |
| Month: | 12 |
| 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. |
| Userfields: | ipdinar={2005}, ipdmembership={speech}, |
| Keywords: | |
| Projects |
Idiap |
| Authors | |
| Added by: | [UNK] |
| Total mark: | 0 |
|
Attachments
|
|
|
|
|
Notes
|
|
|
|
|
