Abstract
In this paper we show the potential of two new features as powerful anti-spoofing measures for face-voice person authentication systems. The features based on latent semantic analysis (LSA) and canonical correlation analysis (CCA), enhance the performance of the authentication system in terms of better anti-imposture abilities and guard against video replay attacks, which is a challenging type of spoof attack. Experiments conducted on 2 speaking-face databases, VidTIMIT and UCBN, show around 42% improvement in error rate with CCA features and 61% improvement with LSA features over feature-level fusion of face-voice feature vectors.
| Original language | English |
|---|---|
| Title of host publication | 6th Interspeech 2005 and 9th European Conference on Speech Communication and Technology (EUROSPEECH) |
| Editors | Trancoso Flurui |
| Place of Publication | Geneva |
| Publisher | International Speech Communication Association |
| Pages | 2181-2184 |
| Number of pages | 4 |
| ISBN (Print) | 9781604234480 |
| Publication status | Published - 2005 |
| Event | 6th Interspeech 2005 and 9th European Conference on Speech Communication and Technology (EUROSPEECH 2005) - Lisboa, Lisboa, Portugal Duration: 4 Sept 2005 → 8 Sept 2005 |
Conference
| Conference | 6th Interspeech 2005 and 9th European Conference on Speech Communication and Technology (EUROSPEECH 2005) |
|---|---|
| Country/Territory | Portugal |
| City | Lisboa |
| Period | 4/09/05 → 8/09/05 |