Intelligent Authentication received a huge boost in the European Union recently, with the announcement that the SpeechXrays consortium, which is supported by EU research and innovation investment, aims to develop and test a multi-channel biometrics solutions based on acoustic and machine vision analysis of speech, lips movement and face.

It is fascinating to observe how technological development continues to catalyse innovations in authentication, together with fears of the same technologies being used to breach these enhanced security systems. While multi-modal biometrics remains a key component of multi-factor authentication (MFA), conventional biometrics, which are largely static in nature, are also evolving to keep pace with the dynamic changes in the world of customer experience and security.

When it comes to voice biometrics, statistical methods are applied to audio acquired from microphones. However, through the use of acoustic imaging sensors, it is also possible to measure musculoskeletal movements that occur while a person is speaking. Examples of these include visible movements such as lips and jaw, but it is these sensors are even pick up subtle movements across a speaker’s face such as eyes and nose using acoustic sensors. This is highly accurate and can be likened to sonar or echolocation that bats and dolphins use for navigation.

The natural next step would be to add facial recognition to this to create a highly accurate, three-modality (voice biometrics, acoustics and facial recognition) to create a highly accurate authentication solution that will most certainly also raise the bar in liveness and anti-spoofing of both voice and facial biometrics.

This is exactly what Idemia, the French ‘Augmented Identity’ company, plans to achieve through the SpeechXrays project. With multi-million-euro funding from the European Union’s Horizon 2020 Research and Innovation Programme, Idemia leads a consortium of partners from across the European Union; and last week welcomed VoiceTrust, the German based Voice and Face Biometrics development company, to the consortium. It is expected that VoiceTrust’s capabilities which span over 15 years of research and product development and deployment will accelerate SpeechXrays’ delivery of a full multi-channel acoustic voice recognition and dynamic face recognition platform.

I will be following the progress on the SpeechXrays project with great interest. The merging of voice biometrics, acoustics and facial recognition will be sure to deliver a step change in Intelligent Authentication.

Categories: Conversational Intelligence, Intelligent Authentication, Articles