Method for processing the output of a speech recognizer
First Claim
1. A method for processing speech, comprising:
- receiving information representing speech input;
parsing the received information representing speech input with respect to a plurality of predetermined command contexts, the plurality of predetermined command contexts comprising command contexts of a plurality of different applications, in an automated speech information processing system;
determining, with at least one automated processor;
if the parsed received information representing speech input corresponds to a command of a respective predetermined command context for execution of the command, and based on the determination that the parsed received information representing speech input corresponds to a command of the respective predetermined command context for reliable execution, then executing the command by a respective application associated with the respective predetermined command context;
if the parsed received information representing speech input cannot be executed, and based on the determination that the parsed received information representing speech input cannot be executed, then prompting for further information representing speech input, to reduce ambiguity or increase completeness, in dependence on previously received information representing speech input and the plurality of predetermined command contexts; and
if an abort, fail or cancel condition is detected in further information representing speech input, and based on the determination that the abort, fail or cancel condition is detected in the further information representing speech input, then exiting said determining.
1 Assignment
0 Petitions
Accused Products
Abstract
A method for processing speech, comprising semantically parsing a received natural language speech input with respect to a plurality of predetermined command grammars in an automated speech processing system; determining if the parsed speech input unambiguously corresponds to a command and is sufficiently complete for reliable processing, then processing the command; if the speech input ambiguously corresponds to a single command or is not sufficiently complete for reliable processing, then prompting a user for further speech input to reduce ambiguity or increase completeness, in dependence on a relationship of previously received speech input and at least one command grammar of the plurality of predetermined command grammars, reparsing the further speech input in conjunction with previously parsed speech input, and iterating as necessary. The system also monitors abort, fail or cancel conditions in the speech input.
324 Citations
23 Claims
-
1. A method for processing speech, comprising:
-
receiving information representing speech input; parsing the received information representing speech input with respect to a plurality of predetermined command contexts, the plurality of predetermined command contexts comprising command contexts of a plurality of different applications, in an automated speech information processing system; determining, with at least one automated processor; if the parsed received information representing speech input corresponds to a command of a respective predetermined command context for execution of the command, and based on the determination that the parsed received information representing speech input corresponds to a command of the respective predetermined command context for reliable execution, then executing the command by a respective application associated with the respective predetermined command context; if the parsed received information representing speech input cannot be executed, and based on the determination that the parsed received information representing speech input cannot be executed, then prompting for further information representing speech input, to reduce ambiguity or increase completeness, in dependence on previously received information representing speech input and the plurality of predetermined command contexts; and if an abort, fail or cancel condition is detected in further information representing speech input, and based on the determination that the abort, fail or cancel condition is detected in the further information representing speech input, then exiting said determining. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10)
-
-
11. A method for processing speech, comprising:
-
receiving speech input; automatically processing the received speech input with at least one automated processor, to parse the speech input, in accordance with a plurality of available command contexts, associated with a plurality of different applications which accept respective commands, and if a correspondence of previously received speech input to any single command is incomplete for execution by any of the plurality of different applications, automatically prompting a human user for further speech input, with a prompt adapted to solicit information from the human user to reduce ambiguity or increase completeness with respect to respective commands accepted by the available command contexts; automatically disregarding previously receive speech input if an abort, fail or cancel condition is detected by the at least one automated processor in further speech input; and based on a result of the automatically processing, selectively processing the command with the at least one automated command processor. - View Dependent Claims (12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22)
-
-
23. A speech processing method, comprising:
-
receiving a speech input; automatically parsing the speech input, in an automated speech information processing system, with respect to a plurality of predetermined command contexts associated with a plurality of different applications which each accept commands for execution; automatically determining; if the parsed speech input represents a command of a respective predetermined command context that can be executed, then executing the command by the respective application which is associated with the respective predetermined command context, else, automatically determining; if the parsed received speech input does not represent a command that can be executed by any respective application which is associated with the plurality of predetermined command contexts, then prompting for further information representing speech input, to reduce ambiguity or increase completeness, in dependence on previously received speech input and the plurality of predetermined command contexts, and automatically determining; if an abort, fail or cancel condition is detected in further speech input, then exiting said determining.
-
Specification