Systems and methods for authentication using sound-based vocalization analysis

A system comprising:

• at least one processor configured to execute software instructions that, upon execution, cause the at least one processor to perform steps to:
• receive a plurality of audio recordings from a user device associated with a user;
• wherein the plurality of audio recordings represent audio from a plurality of forced cough vocalizations produced by the user;
• utilize at least one audio filter to extract a plurality of forced cough vocalization recordings from the plurality of audio recordings;
• generate a plurality of signal data signatures representative of the plurality of forced cough vocalization recordings;
• generate a plurality of combinations of signal data signatures, wherein each combination of signal data signatures of the plurality of combinations of signal data signatures comprises a unique combination of:
• a first group of non-repeated signal data signatures of the plurality of signal data signatures and a second group of non-repeated signal data signatures of the plurality of signal data signatures;
• generate a plurality of Gaussian mixture models, wherein each Gaussian mixture model of the plurality of Gaussian mixture models is formed from the first group of non-repeated signal data signatures of each combination of each combination of signal data signatures of the plurality of combinations of signal data signatures;
• utilize each Gaussian mixture model of the plurality of Gaussian mixture models to produce at least one match value for the second group of non-repeated signal data signatures of each Gaussian mixture model based at least in part on the first group of non-repeated signal data signatures of each Gaussian mixture model;
• wherein the at least one match value of each second group of non-repeated signal data signatures comprises at least one probability of each second group of non-repeated signal data signatures matching to each first group of non-repeated signal data signatures;
• determine at least one statistical score associated with the at least one match value of each Gaussian mixture model;
• determine at least one baseline model from the plurality of Gaussian mixture models based at least in part on the at least one statistical score associated with the at least one match value of each Gaussian mixture model;
• and
• store the at least one baseline model is associated with a user profile associated with the user to enable authentication of at least one subsequent audio recording.

A non-transitory computer readable medium having software instructions stored thereon, the software instructions configured to cause at least one processor to perform steps comprising:

• receiving a plurality of audio recordings from a user device associated with a user;
• wherein the plurality of audio recordings represent audio from a plurality of forced cough vocalizations produced by the user;
• utilizing at least one audio filter to extract a plurality of forced cough vocalization recordings from the plurality of audio recordings;
• generating a plurality of signal data signatures representative of the plurality of forced cough vocalization recordings;
• generating a plurality of combinations of signal data signatures, wherein each combination of signal data signatures of the plurality of combinations of signal data signatures comprises a unique combination of:
• a first group of non-repeated signal data signatures of the plurality of signal data signatures and a second group of non-repeated signal data signatures of the plurality of signal data signatures;
• generating a plurality of Gaussian mixture models, wherein each Gaussian mixture model of the plurality of Gaussian mixture models is formed from the first group of non-repeated signal data signatures of each combination of each combination of signal data signatures of the plurality of combinations of signal data signatures;
• utilizing each Gaussian mixture model of the plurality of Gaussian mixture models to produce at least one match value for the second group of non-repeated signal data signatures of each Gaussian mixture model based at least in part on the first group of non-repeated signal data signatures of each Gaussian mixture model;
• wherein the at least one match value of each second group of non-repeated signal data signatures comprises at least one probability of each second group of non-repeated signal data signatures matching to each first group of non-repeated signal data signatures;
• determining at least one statistical score associated with the at least one match value of each Gaussian mixture model;
• determining at least one baseline model from the plurality of Gaussian mixture models based at least in part on the at least one statistical score associated with the at least one match value of each Gaussian mixture model;
• and
• storing the at least one baseline model is in associated with a user profile associated with the user to enable authentication of at least one subsequent audio recording.