Spectral local harmonicity feature for voice activity detection
In this paper, we propose a method to exploit the harmonicity of human voiced speech using only the most harmonic sub-part of the spectrum. This technique searches for all the potential sub-windows of the spectrum, and measures their local harmonicity, using a newly proposed metric, which works in t...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Conference Paper |
| Language: | English |
| Published: |
2013
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/101329 http://hdl.handle.net/10220/16271 |
| _version_ | 1811682065656053760 |
|---|---|
| author | Khoa, Pham Chau Siong, Chng Eng |
| author2 | School of Computer Engineering |
| author_facet | School of Computer Engineering Khoa, Pham Chau Siong, Chng Eng |
| author_sort | Khoa, Pham Chau |
| collection | NTU |
| description | In this paper, we propose a method to exploit the harmonicity of human voiced speech using only the most harmonic sub-part of the spectrum. This technique searches for all the potential sub-windows of the spectrum, and measures their local harmonicity, using a newly proposed metric, which works in the spectral autocorrelation domain and employs a novel sinusoidal fitting approach. Experiments show that the new feature can be used to detect noisy voiced speech frames heavily corrupted by non-stationary noise even at 0dB SNR with high precision and recall, which gives better results than the Windowed Autocorrelation Lag Energy (WALE), a recently proposed voicing features, under a complex factory noise scenarios. |
| first_indexed | 2024-10-01T03:50:55Z |
| format | Conference Paper |
| id | ntu-10356/101329 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T03:50:55Z |
| publishDate | 2013 |
| record_format | dspace |
| spelling | ntu-10356/1013292020-05-28T07:19:14Z Spectral local harmonicity feature for voice activity detection Khoa, Pham Chau Siong, Chng Eng School of Computer Engineering International Conference on Audio, Language and Image Processing (2012 : Shanghai, China) Temasek Laboratories DRNTU::Engineering::Computer science and engineering In this paper, we propose a method to exploit the harmonicity of human voiced speech using only the most harmonic sub-part of the spectrum. This technique searches for all the potential sub-windows of the spectrum, and measures their local harmonicity, using a newly proposed metric, which works in the spectral autocorrelation domain and employs a novel sinusoidal fitting approach. Experiments show that the new feature can be used to detect noisy voiced speech frames heavily corrupted by non-stationary noise even at 0dB SNR with high precision and recall, which gives better results than the Windowed Autocorrelation Lag Energy (WALE), a recently proposed voicing features, under a complex factory noise scenarios. 2013-10-04T06:29:46Z 2019-12-06T20:36:48Z 2013-10-04T06:29:46Z 2019-12-06T20:36:48Z 2012 2012 Conference Paper Khoa, P. C., & Siong, C. E. (2012). Spectral local harmonicity feature for voice activity detection. 2012 International Conference on Audio, Language and Image Processing (ICALIP). https://hdl.handle.net/10356/101329 http://hdl.handle.net/10220/16271 10.1109/ICALIP.2012.6376652 en |
| spellingShingle | DRNTU::Engineering::Computer science and engineering Khoa, Pham Chau Siong, Chng Eng Spectral local harmonicity feature for voice activity detection |
| title | Spectral local harmonicity feature for voice activity detection |
| title_full | Spectral local harmonicity feature for voice activity detection |
| title_fullStr | Spectral local harmonicity feature for voice activity detection |
| title_full_unstemmed | Spectral local harmonicity feature for voice activity detection |
| title_short | Spectral local harmonicity feature for voice activity detection |
| title_sort | spectral local harmonicity feature for voice activity detection |
| topic | DRNTU::Engineering::Computer science and engineering |
| url | https://hdl.handle.net/10356/101329 http://hdl.handle.net/10220/16271 |
| work_keys_str_mv | AT khoaphamchau spectrallocalharmonicityfeatureforvoiceactivitydetection AT siongchngeng spectrallocalharmonicityfeatureforvoiceactivitydetection |