Privacy preserving biometric signature generation

A computer-implemented method, the method comprising:

• detecting, at an electronic device, one or more biometric image signals of a user from an image sensing device;
• detecting, at the electronic device, one or more raw acoustic signals associated with the user from an acoustic sensing device;
• identifying one or more biometric acoustic signals derived from the one or more raw acoustic signals;
• generating a biometric signal input of the one or more biometric acoustic signals and one or more biometric image signals;
• applying a machine learning machine learning model to conduct feature extraction of the biometric signal input having one or more biometric acoustic signals and one or more biometric image signals;
• generating a biometric user signature of the user from the machine learning model;
• performing a privacy preserving hashing function to the biometric user signature to generate a hashed biometric user signature;
• determining whether the hashed biometric user signature satisfies a predetermined probabilistic threshold with an enrollment hashed signature of the user;
• and
• authenticating an identity of the user upon detecting that the hashed biometric user signature satisfies the predetermined threshold.