Visualization interface for voice input
AssigneeComcast Cable Communications, LLC
InventorsFerhan Ture, Md Iftekhar Tanveer
Life of this patent
Estimating the term…
Abstract
Voice inputs may be analyzed based on syntactic properties and respective dependency tree structure (e.g., parse tree) may be generated. Dependency tree structures may be grouped/clustered based on an associated root word/phrase. A visual interface may display the dependency tree structure in a manner that maps each voice input to a proper response (e.g., an action, an operation, a command, etc.). Mapping voice inputs to responses may be used to train a neural network,
Claims20
A method comprising:
- determining, based on at least a portion of a first voice input received via a voice enabled device, a first root phrase;
- determining, based on the first root phrase and language rules, a first data structure, wherein the first data structure comprises a first plurality of sub-phrases derived from the first voice input;
- generating, based on the first root phrase and the first data structure, a first query;
- determining based on at least a portion of a second voice input received via the voice enabled device, a second root phrase;
- determining, based on the second root phrase and the language rules, a second data structure, wherein the second data structure comprises a second plurality of sub-phrases derived from the second voice input;
- generating, based on the second root phrase and the second data structure, a second query;
- and
- determining, based on at least a portion of the first query and at least a portion of the second query, a data cluster, wherein the data cluster correlates the first query to the second query.
A method comprising:
- receiving, via a voice enabled device, a plurality of voice inputs;
- determining, based on the plurality of voice inputs, a plurality of root phrases, wherein each root phrase of the plurality of root phrases is associated with a voice input of the plurality of voice inputs;
- determining, for each root phrase of the plurality of root phrases, a reference value of a plurality of reference values;
- determining, for each response of a plurality of responses, a correlation value of a plurality of correlation values;
- and
- determining, for each voice input of the plurality of voice inputs, an association between the voice input and one or more responses of the plurality of responses.
A method comprising:
- receiving, by a computing device via a voice enabled device, a first voice input;
- determining, based on at least a portion of the first voice input, a first root phrase;
- determining, based on the first root phrase and language rules, a first data structure;
- receiving, from the voice enabled device, a second voice input;
- determining, based on at least a portion of the second voice input, a second root phrase;
- determining, based on the second root phrase and the language rules, a second data structure;
- and
- providing, to the voice enabled device, the first data structure and the second data structure.
The method of claim 17, wherein determining the data cluster comprises determining an association between the first data structure and the second data structure, wherein the association is based on a match between a definition associated with the first root phrase and a definition associated with the second root phrase satisfying a threshold.
Dependencies are inferred from the claim text.