Dynamic face and voice signature authentication for enhanced security
AssigneeMicrosoft Technology Licensing, LLC
InventorsCarson J. Hicks, John C. Gordon
Life of this patent
Estimating the term…
Abstract
Techniques and apparatuses for a dynamic face and voice signature authentication for enhanced authentication techniques are described herein. In one or more implementations, an authentication system is configured to authenticate a user using a combination of voice and facial recognition techniques. The authentication system supports multiple phrases per user, such that the user can use different phrases to gain different types of access to a device or resources. Doing so provides enhanced customized access to the device or resources.
Claims20
A computer-implemented method for authenticating a user, the method comprising:
- receiving a video stream and an audio stream of a user speaking;
- processing the video stream to identify a face of the user in the video stream;
- generating a dynamic face and voice signature of the user by correlating facial motion of the face of the user in the processed video stream to a voice of the user speaking in the audio stream;
- determining that the dynamic face and voice signature of the user substantially matches a stored dynamic face and voice signature of an authorized user;
- scanning the audio stream to detect an authentication phrase spoken by the user based on the determining that the dynamic face and voice signature of the user substantially matches the stored dynamic face and voice signature;
- and
- authenticating the user and providing access to a device or resources in accordance with the detected authentication phrase in the audio stream.
A system comprising:
- a video capture device configured to capture a video stream of a user speaking:
- a voice capture device configured to capture an audio stream of the user speaking;
- one or more memories and a processing system to implement an authentication system, the authentication system configured to perform operations comprising:
- receiving the video stream of the user speaking from the video capture device, and receiving the audio stream of the user speaking from the audio capture device;
- processing the video stream to identify a face of the user;
- generating a dynamic face and voice signature by correlating facial motion of the face of the user in the processed video stream to a voice of the user speaking in the audio stream;
- determining that the dynamic face and voice signature of the user substantially matches a stored dynamic face and voice signature of an authorized user;
- and
- scanning the audio stream to detect an authentication phrase spoken by the user based on the determining that the dynamic face and voice signature of the user substantially matches the stored dynamic face and voice signature.
Dependencies are inferred from the claim text.