SpeechXRays. Multi-channel biometrics combining acoustic and machine vision analysis of speech, lip movement and face, project funded from the European Union’s Horizon 2020 Research and Innovation Programme, under Grant Agreement no 653386, will develop and test in real-life environments a user recognition platform based on voice acoustics analysis and audio-visual identity verification. SpeechXRays will outperform state-of-the-art solutions with respect to security, privacy, usability, and cost-efficiency.
The project will combine and pilot two proven techniques:
- Acoustic driven voice recognition (using acoustic rather than statistical only models)
- Multi-channel biometrics incorporating dynamic face recognition (machine vision analysis of speech, lip movement and face).
The vision of the SpeechXRays project is to provide a solution combining the convenience and cost-effectiveness of voice biometrics, achieving better accuracies by combining it with video, and bringing superior anti-spoofing capabilities.
The consortium of the SpeechXRays project is headed by IDEMIA France and consists of SIVECO (Romania), Tech Inspire (United Kingdom), Realeyes (Estonia), Forthnet (Greece), IFIN-HH (Romania), FORTH (Greece), University College London (United Kingdom), Institut Mines-Télécom (France), and VoiceTrust (Germany).
Among the peer-reviewed publications of SpeechXRays team of IFIN-HH we mention here:
- A. Cocioceanu, M. Barbulescu, T. Ivanoaica, M. Raportaru, and A.I. Nicolin, Testing
voice-based biometrics authentication platforms for Romanian utterances through
infrequent consonant clusters, Proceedings of the 15th RoEduNet Conference -
Networking in Education and Research, Bucharest, Romania, September 7-9, 2016, DOI:
- A. Cocioceanu, M.C. Raportaru, A.I. Nicolin, and, D. Jakimovski, Rank-frequency
distributions of Romanian words, AIP Conference Proceedings 1916, 040006 (2017),
- A. Cocioceanu, T. Ivanoaica, A.I. Nicolin, and M.C. Raportaru, Computer-based
statistical description of phonetical balance for Romanian utterances, Advances in
Intelligent Systems and Computing 665, 59-67 (2018) DOI: 10.1007/978-3-319-68855-8_6
- A.N. Cocioceanu, S. Micluta-Campeanu, M.C. Raportaru, E.G. Spanakis, and A.I.
Nicolin, Statistical description of phonetic balance for Greek utterances, AIP
Conference Proceedings 2071, 040016 (2019), DOI: 10.1063/1.5090083
Among the other publications we mention the third newsletter of the SpeechXRays project, issued in November 2017, dedicated to data protection regulations, which can be downloaded from here.
The members of the SpeechXRays team of IFIN-HH delivered presentations about the workforce use case and the project at large at the EAB Research Projects Conference in Darmstadt, Germany, at the ICT Innovations conference in Ohrid, Republic of North Macedonia, at the Institute of Physics Belgrade, Serbia, at the Faculty of Computer Science and Engineering, Ss. Cyril and Methodius University in Skopje, Republic of North Macedonia