Automated word correction in speech recognition systems
AssigneeRovi Guides, Inc.
InventorsAnkur Anil Aher, Jeffry Copps Robert Jose
Life of this patent
Estimating the term…
Abstract
Systems and methods for correcting recognition errors in speech recognition systems are disclosed herein. Natural conversational variations are identified to determine whether a query intends to correct a speech recognition error or whether the query is a new command. When the query intends to correct a speech recognition error, the system identifies a location of the error and performs the correction. The corrected query can be presented to the user or be acted upon as a command for the system.
Claims20
A method for correcting a speech recognition error, the method comprising:
- detecting a first voice input;
- detecting a second voice input within a threshold time of detecting the first voice input;
- and
- in response to detecting the second voice input within the threshold time of detecting the first voice input, determining whether the second voice input is directed to correcting the first voice input based on:
- comparing a first plurality of sound properties of the first voice input to a second plurality of sound properties of the second voice input;
- and
- determining, based on the comparing, that the second voice input is directed to correcting the first voice input when at least a threshold amount of the first plurality of sound properties match the second plurality of sound properties and at least one sound property of the first plurality of sound properties does not match the second plurality of sound properties.
A system for correcting a speech recognition error, the system comprising control circuitry configured to:
- detect a first voice input;
- detect a second voice input within a threshold time of detecting the first voice input;
- and
- in response to detecting the second voice input within the threshold time of detecting the first voice input, determine whether the second voice input is directed to correcting the first voice input based on:
- comparing a first plurality of sound properties of the first voice input to a second plurality of sound properties of the second voice input;
- and
- determining, based on the comparing, that the second voice input is directed to correcting the first voice input when at least a threshold amount of the first plurality of sound properties match the second plurality of sound properties and at least one sound property of the first plurality of sound properties does not match the second plurality of sound properties.
Dependencies are inferred from the claim text.