Noise compensation in a person verification system using face and multiple speech features

Pattern Recognition

Volumn 36, Issue 2, 2003, Pages 293-302

  Sanderson, Conrad ^a   Paliwal, Kuldip K ^a  

^a GRIFFITH UNIVERSITY   (Australia)

Author keywords

Adaptive fusion; Audio visual; Identity verification; Multi modal; Source features

Indexed keywords

AUDIO ACOUSTICS; DATABASE SYSTEMS; ROBUSTNESS (CONTROL SYSTEMS);

PERSON VERIFICATION SYSTEM;

SPEECH;

PATTERN RECOGNITION;

EID: 0036487270     PISSN: 00313203     EISSN: None     Source Type: Journal    
DOI: 10.1016/S0031-3203(02)00031-6     Document Type: Article

References (28)

1. Speaker identification and verification using Gaussian mixture speaker models
2. On the use of instantaneous and transitional spectral information in speaker recognition
3. Cepstral analysis technique for automatic speaker verification
4. Use of voicing and pitch information for speaker recognition
5. Integrating faces and fingerprints for personal identification
6. Information fusion in biometrics
7. Fusion off ace and speech data for person identity verification
8. Frequency analysis and masking
9. Signal modeling techniques in speech recognition
10. Channel estimation and normalization by coherent spectral averaging for robust speaker verification
11. Text-independent speaker identification
12. On the use of autocorrelation analysis for pitch detection
13. Eigenfaces for recognition
14. Expectation-maximization algorithm
15. Recent advances in speaker recognition
16. Computer lipreading for improved accuracy in automatic speech recognition
17. Adaptive multi-modal person verification system
18. NTIMIT: A phonetically balanced, continuous speech telephone bandwidth speech database
19. Speaker Verification using adapted Gaussian mixture models
20. Why recognition in a statistics-based face recognition system should be based on the pure face portion: A probabilistic decision-based proof