Efficient Invariant Features for Sensor Variability Compensation in Speaker Recognition
In this paper, we investigate the use of invariant features for speaker recognition. Owing to their characteristics, these features are introduced to cope with the difficult and challenging problem of sensor variability and the source of performance degradation inherent in speaker recognition system...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2014-10-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | http://www.mdpi.com/1424-8220/14/10/19007 |
| _version_ | 1818012780933939200 |
|---|---|
| author | Abdennour Alimohad Ahmed Bouridane Abderrezak Guessoum |
| author_facet | Abdennour Alimohad Ahmed Bouridane Abderrezak Guessoum |
| author_sort | Abdennour Alimohad |
| collection | DOAJ |
| description | In this paper, we investigate the use of invariant features for speaker recognition. Owing to their characteristics, these features are introduced to cope with the difficult and challenging problem of sensor variability and the source of performance degradation inherent in speaker recognition systems. Our experiments show: (1) the effectiveness of these features in match cases; (2) the benefit of combining these features with the mel frequency cepstral coefficients to exploit their discrimination power under uncontrolled conditions (mismatch cases). Consequently, the proposed invariant features result in a performance improvement as demonstrated by a reduction in the equal error rate and the minimum decision cost function compared to the GMM-UBM speaker recognition systems based on MFCC features. |
| first_indexed | 2024-04-14T06:24:43Z |
| format | Article |
| id | doaj.art-d94b5a68ff9748c09f6be0194e855524 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-04-14T06:24:43Z |
| publishDate | 2014-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-d94b5a68ff9748c09f6be0194e8555242022-12-22T02:07:53ZengMDPI AGSensors1424-82202014-10-011410190071902210.3390/s141019007s141019007Efficient Invariant Features for Sensor Variability Compensation in Speaker RecognitionAbdennour Alimohad0Ahmed Bouridane1Abderrezak Guessoum2Research Laboratory in Electrical Engineering and Automatic LREA, University of MEDEA, Ain D'heb, Medea 26000, AlgeriaSchool of Computing, Engineering and Information Sciences, Northumbria University, Newcastle Upon Tyne NE2 1XE, UKDepartment of Electronics Engineering, University of Blida, Blida BP 270, AlgeriaIn this paper, we investigate the use of invariant features for speaker recognition. Owing to their characteristics, these features are introduced to cope with the difficult and challenging problem of sensor variability and the source of performance degradation inherent in speaker recognition systems. Our experiments show: (1) the effectiveness of these features in match cases; (2) the benefit of combining these features with the mel frequency cepstral coefficients to exploit their discrimination power under uncontrolled conditions (mismatch cases). Consequently, the proposed invariant features result in a performance improvement as demonstrated by a reduction in the equal error rate and the minimum decision cost function compared to the GMM-UBM speaker recognition systems based on MFCC features.http://www.mdpi.com/1424-8220/14/10/19007speaker recognitioninvariant featuresMFCCsGMM-UBMsensor variabilityDET curve |
| spellingShingle | Abdennour Alimohad Ahmed Bouridane Abderrezak Guessoum Efficient Invariant Features for Sensor Variability Compensation in Speaker Recognition Sensors speaker recognition invariant features MFCCs GMM-UBM sensor variability DET curve |
| title | Efficient Invariant Features for Sensor Variability Compensation in Speaker Recognition |
| title_full | Efficient Invariant Features for Sensor Variability Compensation in Speaker Recognition |
| title_fullStr | Efficient Invariant Features for Sensor Variability Compensation in Speaker Recognition |
| title_full_unstemmed | Efficient Invariant Features for Sensor Variability Compensation in Speaker Recognition |
| title_short | Efficient Invariant Features for Sensor Variability Compensation in Speaker Recognition |
| title_sort | efficient invariant features for sensor variability compensation in speaker recognition |
| topic | speaker recognition invariant features MFCCs GMM-UBM sensor variability DET curve |
| url | http://www.mdpi.com/1424-8220/14/10/19007 |
| work_keys_str_mv | AT abdennouralimohad efficientinvariantfeaturesforsensorvariabilitycompensationinspeakerrecognition AT ahmedbouridane efficientinvariantfeaturesforsensorvariabilitycompensationinspeakerrecognition AT abderrezakguessoum efficientinvariantfeaturesforsensorvariabilitycompensationinspeakerrecognition |