ARTICLE jourlin-prl97/IDIAP Acoustic-Labial Speaker Verification Jourlin, Pierre Luettin, Juergen Genoud, Dominique Wassner, H. EXTERNAL https://publications.idiap.ch/attachments/papers/1997/prl97.pdf PUBLIC https://publications.idiap.ch/index.php/publications/showcite/jourlin-rr-97-13 Related documents Pattern Recognition Letters 18 09 853-858 1997 IDIAP-RR 97-13 This paper describes a multimodal approach for speaker verification. The system consists of two classifiers, one using visual features and the other using acoustic features. A lip tracker is used to extract visual information from the speaking face which provides shape and intensity features. We describe an approach for normalizing and mapping different modalities onto a common confidence interval. We also describe a novel method for integrating the scores of multiple classifiers. Verification experiments are reported for the individual modalities and for the combined classifier. The performance of the integrated system outperformed each sub-system and reduced the false acceptance rate of the acoustic sub-system from 2.3\% to 0.5\%. REPORT jourlin-RR-97-13/IDIAP Acoustic-Labial Speaker Verification Jourlin, Pierre Luettin, Juergen Genoud, Dominique Wassner, H. EXTERNAL https://publications.idiap.ch/attachments/reports/1997/rr97-13.pdf PUBLIC Idiap-RR-13-1997 1997 IDIAP To appear in Pattern Recognition Letters This paper describes a multimodal approach for speaker verification. The system consists of two classifiers, one using visual features and the other using acoustic features. A lip tracker is used to extract visual information from the speaking face which provides shape and intensity features. We describe an approach for normalizing and mapping different modalities onto a common confidence interval. We also describe a novel method for integrating the scores of multiple classifiers. Verification experiments are reported for the individual modalities and for the combined classifier. The performance of the integrated system outperformed each sub-system and reduced the false acceptance rate of the acoustic sub-system from 2.3\% to 0.5\%.