Bank of America Corporation

Swetha Ravi

Pratap Dande

Kevin A Delson

Lalit Dhawan

Morgan Stewart Allen

Gilbert M Gatchalian

Rohit Sawkar

Apparatus and methods are provided for an autonomous bot that learns from user activity. The bot then autonomously acts to complete tasks that the user typically performs. The bot may detect and act in response to detecting a problem with an application. The bot may dynamically generate guidance that assists the user completing a target task. Illustrative guidance may include highlighting differences between software applications, identifying information missing from a document or form and demonstrating how to obtain the needed information or complete a task.

Self-healing bot

Systems and methods for multi-dimensional voice-based digital authentication are provided. Methods include receiving, at a central server from a remote user device, a request to access a protected digital resource; displaying a prompt, at the device, requesting a spoken password expression; capturing, at the device, an expression spoken in response to the prompt; determining that the expression satisfies predetermined attributes of the password expression; calculating a voice pattern representation of the expression; determining that the voice pattern representation of the expression satisfies a threshold similarity to a voice pattern representation of the user that is stored in a database at the central server; and, in response to the expression satisfying the predetermined attributes of the password expression and satisfying the threshold similarity to the voice pattern representation of the user that is stored in the database at the central server, authorizing the device to access the protected digital resource.

Multi-dimensional voice-based digital authentication

Bots are typically programmed to automate tasks and provide statistically expected results. However, a bot may malfunction and generate aberrant outputs. It is technically challenging to detect the aberrant outputs and determine whether the outputs are due to an error in how the bot processes inputs or because the bot has received unusual or unexpected input data. Apparatus and methods are provided for auto-determining why a bot has generated output outside expected results. An auto-correct bot will detect problems and auto-identify potential solutions, simulate those solutions and apply those solutions to remediate the malfunctioning bot.

Growing community of inventors

Charlotte, NC, United States of America

Swetha Ravi