Leveraging Noise Cancelling Technology for Dynamic Voice Authentication

A method, comprising:

• receiving, by a computing platform, the computing platform having at least one processor, and memory, audio data from a user, wherein the audio data is captured by a mobile device of the user and the audio data is received via the mobile device of the user;
• initiating, by the at least one processor and based on the received audio data, a transaction session;
• executing, by the at least one processor, sound cancelling techniques to isolate the audio data;
• comparing, by the at least one processor, the audio data to pre-stored user authentication data to determine whether the user is authenticated to the transaction session;
• responsive to determining that the user is not authenticated to the transaction session, terminating, by the at least one processor, the transaction session;
• responsive to determining that the user is authenticated to the transaction session:
• extracting, by the at least one processor and from the audio data, features, wherein extracting the features results in an audio signal formatted for further processing;
• executing, by the at least one processor, one or more speech recognition techniques on the audio signal to generate a plurality of phonetic units;
• executing, by the at least one processor, one or more machine learning models, wherein executing the one or more machine learning models includes inputting, to the one or more machine learning models, the plurality of phonetic units to generate an output;
• and
• transmitting, by the at least one processor and to the mobile device of the user, the generated output for confirmation.

One or more non-transitory computer-readable media storing instructions that, when executed by a computing platform comprising at least one processor, memory, and a communication interface, cause the computing platform to:

• receive audio data from a user, wherein the audio data is captured by a mobile device of the user and the audio data is received via the mobile device of the user;
• initiate, based on the received audio data, a transaction session;
• execute sound cancelling techniques to isolate the audio data;
• compare the audio data to pre-stored user authentication data to determine whether the user is authenticated to the transaction session;
• responsive to determining that the user is not authenticated to the transaction session, terminate the transaction session;
• responsive to determining that the user is authenticated to the transaction session:
• extract, from the audio data, features, wherein extracting the features results in an audio signal formatted for further processing;
• execute one or more speech recognition techniques on the audio signal to generate a plurality of phonetic units;
• execute one or more machine learning models, wherein executing the one or more machine learning models includes inputting, to the one or more machine learning models, the plurality of phonetic units to generate an output;
• and
• transmit, to the mobile device of the user, the generated output for confirmation.