SU SecurePhone home page

project logo logo SU
Secure contracts signed by mobile phone

Funding : The project (IST-2002-506883) is funded in the European Union's 6th Framework Programme with 2 mln Euros.

Duration : The project runs for 30 months, January 2004 - June 2006.

Partners : group photo

Atos Origin (AO) (E)
Informa (IN) (E)
Telefónica Móviles España (TM) (E)
Nergal (NE) (I)
The Institute of Phonetics of the University of Saarbrücken (SU) (D)
The University of Buckingham (BU) (UK)
Group des Écoles des Télécommunications - INT (GI ) (F)
Group des Écoles des Télécommunications - ENST (GE ) (F)

Project aim
The SecurePhone project integrates a biometric recogniser in a 3G/B3G enabled PDA, which allows users to mutually recognise each other and securely authenticate messages (text or audio). This will enable them to sign legally binding contracts on their PDA/mobile phone. The biometric recogniser combines source authentication methods on the basis of text-dependent speaker verification, video recordings of the speaker's face and a written signature. More information will be made available later.

Local project web pages
here

Saarbrücken University (SU) role in project
The main responsibility of Saarland University is text-dependent speaker verification and coordination of multimodal data fusion (speech with face image and handwritten signature). Our initial project aims are to compare HMM with HMM or DTW of parameter trajectories in a Kohonen network (calibrated with speakers). We shall also explore the potential use of wavelets in the preprocessing stage, in place of the standard MFCCs.

SU collaborators
Jacques Koreman homepage Email
Andrew C Morris homepage Email
Dalei Wu homepage Email

We also work together with the Department for Spoken Language Systems

You can find more information on the main project webpage:
SecurePhone home page

Publications

Koreman, J., Morris, A.C., Jassim, S., Sellahewa, H., Chollet, G., Aversano, G., Salicetti, S. & Allano, L. (2006) "Multi-modal biometric authentication on the SecurePhone PDA", Proc. MMUA workshop on Mult-Modal User Authentication, Toulouse, France, May 2006. Pdf
Morris, A.C., Jassim, S., Sellahewa, H., Allano, L., Ehlers, J., Wu, D., Koreman, J., Garcia-Salicetti, S., Ly-Vav, B. & Dorizzi, B. (2006) "Multimodal person authentication on a smartphone under realistic conditions", Proc. SPIE conference on mobile multimedia/image processing for military and security applications, Orlando (in press). Pdf
Morris, A.C., Koreman, J., Sellahewa, H. Ehlers, J., Jassim, S., Allano, L. & Garcia-Salicetti, S., (2006) "The SecurePhone PDA database, experimental protocol and automatic test procedure for multi-modal user authentication", Tech. Report, Saarland University , Institute of Phonetics. Pdf
Morris, A.C., Wu, D. & Koreman, J. (2005) "MLP trained to separate problem speakers provides improved features for speaker identification", Proc. ICCST 2005. Pdf
Wu, D., Morris, A.C. & Koreman, J. (2006) "SOM trajectory as complementary speech representation in multi-stream speaker recognition", internal document. Pdf

Wu, D., Morris, A.C. & Koreman, J. (2005) "MLP internal representation as discriminative features for improved speaker recognition", Proc. NOLISP 2005. Pdf

Wu, D., Koreman, J. & Morris, A.C. (2005) "Discriminative features by MLP preprocessing for robust speaker recognition in noise", Proc. ESSV 2005 Pdf