Abstract: Various embodiments of the present disclosure provide for detecting audio deepfakes through acoustic prosodic modeling. In one example, an embodiment provides for extracting one or more prosodic features from an audio sample and classifying the audio sample as a deepfake audio sample or an organic audio sample by applying a machine learning model to the one or more prosodic features. The one or more prosodic features can be indicative of one or more prosodic characteristics associated with human speech. Additionally, the machine learning model can be configured as a classification-based detector for audio deepfakes.

Abstract: Various methods are provided for secure two-factor authentication, and more specifically, for incorporating a layer of security to two-factor authentication using Short Message Service in a manner virtually transparent to the end-user. Methods may include receiving a request for registration for two-factor authentication from a client including a username and password; providing a request for a mobile device number; receiving the mobile device number and a pre-shared key; sending to a mobile device an identity of the client and a server key share; receiving from the mobile device a mobile device key share; sending information corresponding to an exchange with the mobile device and a challenge derived from the pre-shared key to the client in response to the device key share corresponding to the server key share; receiving confirmation of registration with the mobile device; and establishing a shared key in response to verification of the confirmation.