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:
• control a display of a computing device to render a forced cough vocalization graphical user interface (GUI) comprising at least one user instruction instructing the user to produce a forced cough vocalization;
• control a recording device associated with the computing device to record audio;
• receive, from the recording device, at least one audio signal that captures the forced cough vocalization produced by the user for auscultation;
• utilize at least one pre-processing algorithm to pre-process the at least one audio signal and output forced cough vocalization signal data representative of the forced cough vocalization;
• utilize at least one cough signature model to ingest the forced cough vocalization signal data and output a signal data signature representative of isolated forced cough vocalization-related signal data, wherein the isolated forced cough vocalization-related signal data is isolated from non-forced cough vocalization-related signal data;
• wherein the at least one cough signature model to ingests the forced cough vocalization signal data and outputs the signal data signature based at least in part on at least one machine learning layer comprising a plurality of machine learning parameters;
• wherein the plurality of machine learning parameters is iteratively refined based on error between:
• previous predictions based on historical signal data signatures and
• known signal data signatures paired with the historical signal data signatures;
• determine a signal data signature invalidation based at least in part on testing of the signal data signature against at least one signal data signature criteria resulting in the signal data signature failing to achieve the at least one signal data signature criteria;
• delete based on the signal data signature invalidation, the signal data signature;
• and
• control the display of the computing device to render the forced cough vocalization GUI comprising at least one subsequent user instruction instructing the user to auscultate and produce a new forced cough vocalization in response to the signal data signature invalidation.

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:

• control a display of a computing device to render a forced cough vocalization graphical user interface (GUI) comprising at least one user instruction instructing the user to auscultate and produce a forced cough vocalization;
• control a recording device associated with the computing device to record audio;
• receive, from the recording device, at least one audio signal that captures the forced cough vocalization produced by the user;
• utilize at least one pre-processing algorithm to pre-process the at least one audio signal and output forced cough vocalization signal data representative of the forced cough vocalization;
• utilize at least one cough signature model to ingest the forced cough vocalization signal data and output a signal data signature representative of isolated forced cough vocalization-related signal data, wherein the isolated forced cough vocalization-related signal data is isolated from non-forced cough vocalization-related signal data;
• wherein the at least one cough signature model to ingests the forced cough vocalization signal data and outputs the signal data signature based at least in part on at least one machine learning layer comprising a plurality of machine learning parameters;
• wherein the plurality of machine learning parameters is iteratively refined based on error between:
• previous predictions based on historical signal data signatures and 56 known signal data signatures paired with the historical signal data signatures;
• determine a signal data signature invalidation based at least in part on testing of the signal data signature against at least one signal data signature criteria resulting in the signal data signature failing to achieve the at least one signal data signature criteria;
• delete based on the signal data signature invalidation, the signal data signature;
• and
• control the display of the computing device to render the forced cough vocalization GUI comprising at least one subsequent user instruction instructing the user to auscultate and produce a new forced cough vocalization in response to the signal data signature invalidation.