Innovative Method for Unsupervised Voice Activity Detection and Classification of Audio Segments
An accurate and noise-robust voice activity detection (VAD) system can be widely used for emerging speech technologies in the fields of audio forensics, wireless communication, and speech recognition. However, in real-life application, the sufficient amount of data or human-annotated data to train s...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2018-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/8290827/ |
| _version_ | 1818558655579029504 |
|---|---|
| author | Zulfiqar Ali Muhammad Talha |
| author_facet | Zulfiqar Ali Muhammad Talha |
| author_sort | Zulfiqar Ali |
| collection | DOAJ |
| description | An accurate and noise-robust voice activity detection (VAD) system can be widely used for emerging speech technologies in the fields of audio forensics, wireless communication, and speech recognition. However, in real-life application, the sufficient amount of data or human-annotated data to train such a system may not be available. Therefore, a supervised system for VAD cannot be used in such situations. In this paper, an unsupervised method for VAD is proposed to label the segments of speech-presence and speech-absence in an audio. To make the proposed method efficient and computationally fast, it is implemented by using long-term features that are computed by using the Katz algorithm of fractal dimension estimation. Two databases of different languages are used to evaluate the performance of the proposed method. The first is Texas Instruments Massachusetts Institute of Technology (TIMIT) database, and the second is the King Saud University (KSU) Arabic speech database. The language of TIMIT is English, while the language of the KSU speech database is Arabic. TIMIT is recorded in only one environment, whereas the KSU speech database is recorded in distinct environments using various recording systems that contain sound cards of different qualities and models. The evaluation of the proposed method suggested that it labels voiced and unvoiced segments reliably in both clean and noisy audio. |
| first_indexed | 2024-12-14T00:15:07Z |
| format | Article |
| id | doaj.art-381f37b9b0594c5a89ea04451dba558f |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-14T00:15:07Z |
| publishDate | 2018-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-381f37b9b0594c5a89ea04451dba558f2022-12-21T23:25:35ZengIEEEIEEE Access2169-35362018-01-016154941550410.1109/ACCESS.2018.28058458290827Innovative Method for Unsupervised Voice Activity Detection and Classification of Audio SegmentsZulfiqar Ali0https://orcid.org/0000-0002-1599-1287Muhammad Talha1https://orcid.org/0000-0002-4246-2524Department of Computer Engineering, College of Computer and Information Sciences, King Saud University, Riyadh, Saudi ArabiaDeanship of Scientific Research, King Saud University, Riyadh, Saudi ArabiaAn accurate and noise-robust voice activity detection (VAD) system can be widely used for emerging speech technologies in the fields of audio forensics, wireless communication, and speech recognition. However, in real-life application, the sufficient amount of data or human-annotated data to train such a system may not be available. Therefore, a supervised system for VAD cannot be used in such situations. In this paper, an unsupervised method for VAD is proposed to label the segments of speech-presence and speech-absence in an audio. To make the proposed method efficient and computationally fast, it is implemented by using long-term features that are computed by using the Katz algorithm of fractal dimension estimation. Two databases of different languages are used to evaluate the performance of the proposed method. The first is Texas Instruments Massachusetts Institute of Technology (TIMIT) database, and the second is the King Saud University (KSU) Arabic speech database. The language of TIMIT is English, while the language of the KSU speech database is Arabic. TIMIT is recorded in only one environment, whereas the KSU speech database is recorded in distinct environments using various recording systems that contain sound cards of different qualities and models. The evaluation of the proposed method suggested that it labels voiced and unvoiced segments reliably in both clean and noisy audio.https://ieeexplore.ieee.org/document/8290827/Voiced and unvoiced segmentationfractal dimensionKatz algorithmTIMIT databaseKSU speech database |
| spellingShingle | Zulfiqar Ali Muhammad Talha Innovative Method for Unsupervised Voice Activity Detection and Classification of Audio Segments IEEE Access Voiced and unvoiced segmentation fractal dimension Katz algorithm TIMIT database KSU speech database |
| title | Innovative Method for Unsupervised Voice Activity Detection and Classification of Audio Segments |
| title_full | Innovative Method for Unsupervised Voice Activity Detection and Classification of Audio Segments |
| title_fullStr | Innovative Method for Unsupervised Voice Activity Detection and Classification of Audio Segments |
| title_full_unstemmed | Innovative Method for Unsupervised Voice Activity Detection and Classification of Audio Segments |
| title_short | Innovative Method for Unsupervised Voice Activity Detection and Classification of Audio Segments |
| title_sort | innovative method for unsupervised voice activity detection and classification of audio segments |
| topic | Voiced and unvoiced segmentation fractal dimension Katz algorithm TIMIT database KSU speech database |
| url | https://ieeexplore.ieee.org/document/8290827/ |
| work_keys_str_mv | AT zulfiqarali innovativemethodforunsupervisedvoiceactivitydetectionandclassificationofaudiosegments AT muhammadtalha innovativemethodforunsupervisedvoiceactivitydetectionandclassificationofaudiosegments |