A speaker verification system for Arabic
In this thesis a new Speaker Verification System (SVS) methodology is developed. The objective is to develop a SVS system that achieves high identification accuracy with low storage requirements for the use in smart card-based systems. This proposed system combines the advantages of the simple struc...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | masterThesis |
| Published: |
2008
|
| Subjects: | |
| Online Access: | https://eprints.kfupm.edu.sa/id/eprint/2522/1/Theses-A_speaker_verification_system_for_Arabic.doc |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1864513382450724864 |
|---|---|
| author | Al-Telmisani,, Essam Abed |
| author2 | unknown |
| author2_role | author |
| author_facet | Al-Telmisani,, Essam Abed unknown |
| author_role | author |
| dc.creator.none.fl_str_mv | Al-Telmisani,, Essam Abed unknown |
| dc.date.none.fl_str_mv | 2008-01 2020 |
| dc.format.none.fl_str_mv | application/msword |
| dc.identifier.none.fl_str_mv | https://eprints.kfupm.edu.sa/id/eprint/2522/1/Theses-A_speaker_verification_system_for_Arabic.doc (2008) A speaker verification system for Arabic. Masters thesis, KingFahd UNiversity of Petroleum and Minerals. |
| dc.language.none.fl_str_mv | en |
| dc.relation.none.fl_str_mv | https://eprints.kfupm.edu.sa/id/eprint/2522/ |
| dc.rights.*.fl_str_mv | info:eu-repo/semantics/openAccess |
| dc.subject.none.fl_str_mv | Systems |
| dc.title.none.fl_str_mv | A speaker verification system for Arabic |
| dc.type.none.fl_str_mv | Thesis NonPeerReviewed info:eu-repo/semantics/publishedVersion info:eu-repo/semantics/masterThesis |
| description | In this thesis a new Speaker Verification System (SVS) methodology is developed. The objective is to develop a SVS system that achieves high identification accuracy with low storage requirements for the use in smart card-based systems. This proposed system combines the advantages of the simple structure of text-dependent systems and the performance of text-independent systems. The feature extraction utilized the Mel-Frequency Cepstrum Coefficients MFCC for its superior performance, and user template is based on a Vector Quantization codebook for fast and efficient pattern classification. A new adaptive threshold is also proposed to achieve improved performance with time and to mitigate temporarily changes in speaker voice. The system is trained and tuned for Arabic speakers using initially certain parts from the holy Qura'an. A sample of eighteen Arabic speakers is used to evaluate the system performance. The system achieves an accuracy of 0.01 for the False Acceptance Rate (FAR), and 0.04 for the False Rejection Rate (FRR). |
| eu_rights_str_mv | openAccess |
| format | masterThesis |
| id | KFUPM_8d81502db539b2ee26708e3f32f94d05 |
| identifier_str_mv | (2008) A speaker verification system for Arabic. Masters thesis, KingFahd UNiversity of Petroleum and Minerals. |
| language_invalid_str_mv | en |
| network_acronym_str | KFUPM |
| network_name_str | King Fahd University of Petroleum and Minerals |
| oai_identifier_str | oai::2522 |
| publishDate | 2008 |
| repository.mail.fl_str_mv | |
| repository.name.fl_str_mv | |
| repository_id_str | |
| spelling | A speaker verification system for ArabicAl-Telmisani,, Essam AbedunknownSystemsIn this thesis a new Speaker Verification System (SVS) methodology is developed. The objective is to develop a SVS system that achieves high identification accuracy with low storage requirements for the use in smart card-based systems. This proposed system combines the advantages of the simple structure of text-dependent systems and the performance of text-independent systems. The feature extraction utilized the Mel-Frequency Cepstrum Coefficients MFCC for its superior performance, and user template is based on a Vector Quantization codebook for fast and efficient pattern classification. A new adaptive threshold is also proposed to achieve improved performance with time and to mitigate temporarily changes in speaker voice. The system is trained and tuned for Arabic speakers using initially certain parts from the holy Qura'an. A sample of eighteen Arabic speakers is used to evaluate the system performance. The system achieves an accuracy of 0.01 for the False Acceptance Rate (FAR), and 0.04 for the False Rejection Rate (FRR).2008-012020ThesisNonPeerReviewedinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesisapplication/mswordhttps://eprints.kfupm.edu.sa/id/eprint/2522/1/Theses-A_speaker_verification_system_for_Arabic.doc (2008) A speaker verification system for Arabic. Masters thesis, KingFahd UNiversity of Petroleum and Minerals. enhttps://eprints.kfupm.edu.sa/id/eprint/2522/info:eu-repo/semantics/openAccessoai::25222019-11-01T13:44:37Z |
| spellingShingle | A speaker verification system for Arabic Al-Telmisani,, Essam Abed Systems |
| status_str | publishedVersion |
| title | A speaker verification system for Arabic |
| title_full | A speaker verification system for Arabic |
| title_fullStr | A speaker verification system for Arabic |
| title_full_unstemmed | A speaker verification system for Arabic |
| title_short | A speaker verification system for Arabic |
| title_sort | A speaker verification system for Arabic |
| topic | Systems |
| url | https://eprints.kfupm.edu.sa/id/eprint/2522/1/Theses-A_speaker_verification_system_for_Arabic.doc |