Statistical voice activity detection using low-variance spectrum estimation and an adaptive threshold
Traditionally voice activity detection algorithms are based on any combination of general speech properties such as temporal energy variations, periodicity, and spectrum. This paper describes a novel statistical method for voice activity detection using a signal to noise ratio measure. The method e...
| Main Authors: | , , |
|---|---|
| Format: | Journal Article |
| Published: |
IEEE Signal Processing Society
2006
|
| Subjects: | |
| Online Access: | http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=1597247 http://hdl.handle.net/20.500.11937/37889 |
| _version_ | 1848755171461431296 |
|---|---|
| author | Davis, A. Nordholm, Sven Togneri, R. |
| author_facet | Davis, A. Nordholm, Sven Togneri, R. |
| author_sort | Davis, A. |
| building | Curtin Institutional Repository |
| collection | Online Access |
| description | Traditionally voice activity detection algorithms are based on any combination of general speech properties such as temporal energy variations, periodicity, and spectrum. This paper describes a novel statistical method for voice activity detection using a signal to noise ratio measure. The method employs a low-variance spectrum estimate and determines an optimal threshold based on the estimated noise statistics. A possible implementation is presented and evaluated over a large test set and compared to current modern standardized algorithms. The evaluations indicate promising results with the proposed scheme being comparable or favorable over the whole test set. |
| first_indexed | 2025-11-14T08:52:04Z |
| format | Journal Article |
| id | curtin-20.500.11937-37889 |
| institution | Curtin University Malaysia |
| institution_category | Local University |
| last_indexed | 2025-11-14T08:52:04Z |
| publishDate | 2006 |
| publisher | IEEE Signal Processing Society |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | curtin-20.500.11937-378892017-01-30T14:09:29Z Statistical voice activity detection using low-variance spectrum estimation and an adaptive threshold Davis, A. Nordholm, Sven Togneri, R. voice activity detector - VAD Voice activity detection statistical decision adaptive voice activity detection Traditionally voice activity detection algorithms are based on any combination of general speech properties such as temporal energy variations, periodicity, and spectrum. This paper describes a novel statistical method for voice activity detection using a signal to noise ratio measure. The method employs a low-variance spectrum estimate and determines an optimal threshold based on the estimated noise statistics. A possible implementation is presented and evaluated over a large test set and compared to current modern standardized algorithms. The evaluations indicate promising results with the proposed scheme being comparable or favorable over the whole test set. 2006 Journal Article http://hdl.handle.net/20.500.11937/37889 http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=1597247 IEEE Signal Processing Society restricted |
| spellingShingle | voice activity detector - VAD Voice activity detection statistical decision adaptive voice activity detection Davis, A. Nordholm, Sven Togneri, R. Statistical voice activity detection using low-variance spectrum estimation and an adaptive threshold |
| title | Statistical voice activity detection using low-variance spectrum estimation and an adaptive threshold |
| title_full | Statistical voice activity detection using low-variance spectrum estimation and an adaptive threshold |
| title_fullStr | Statistical voice activity detection using low-variance spectrum estimation and an adaptive threshold |
| title_full_unstemmed | Statistical voice activity detection using low-variance spectrum estimation and an adaptive threshold |
| title_short | Statistical voice activity detection using low-variance spectrum estimation and an adaptive threshold |
| title_sort | statistical voice activity detection using low-variance spectrum estimation and an adaptive threshold |
| topic | voice activity detector - VAD Voice activity detection statistical decision adaptive voice activity detection |
| url | http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=1597247 http://hdl.handle.net/20.500.11937/37889 |