Method for processing the output of a speech recognizer
First Claim
1. A method for processing speech, comprising:
- receiving an input represent human speech;
processing the received input within alternate concurrent instances of a virtual processing environment executing on at least one automated data processing system;
automatically communicating, between respective alternate instances of the virtual processing environment, at least one status message to coordinate a reduction in a number of the alternate instances of the virtual processing environment;
analyzing;
(1) the received input to determine a human-initiated command;
(2) the received input to determine data associated with the human initiated command;
(3) a completeness and unambiguity of the received input with respect to an ability to at least partially process the human-initiated command; and
if the received input is complete and unambiguous with respect to the ability to at least partially process the human initiated command, commencing processing, andif the received input is incomplete or ambiguous with respect to the ability to at least partially process the human initiated command, prompting for additional input.
0 Assignments
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.
325 Citations
22 Claims
-
1. A method for processing speech, comprising:
-
receiving an input represent human speech; processing the received input within alternate concurrent instances of a virtual processing environment executing on at least one automated data processing system; automatically communicating, between respective alternate instances of the virtual processing environment, at least one status message to coordinate a reduction in a number of the alternate instances of the virtual processing environment; analyzing; (1) the received input to determine a human-initiated command; (2) the received input to determine data associated with the human initiated command; (3) a completeness and unambiguity of the received input with respect to an ability to at least partially process the human-initiated command; and if the received input is complete and unambiguous with respect to the ability to at least partially process the human initiated command, commencing processing, and if the received input is incomplete or ambiguous with respect to the ability to at least partially process the human initiated command, prompting for additional input. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12)
-
-
13. A method, comprising:
-
hosting, in an automated data processing system, at least one virtual workspace configured to support instances of a virtual processing environment; processing the received input within alternate concurrent instances of the virtual processing environment executing on at least one automated data processing system; automatically communicating, between respective alternate instances of the virtual processing environment, at least one status message to coordinate a reduction in a number of the alternate instances of the virtual processing environment; analyzing a completeness and unambiguity of the received input with respect to an ability to at least partially process a human-initiated command; if the received input is complete and unambiguous with respect to the ability to at least partially process the human initiated command, commencing processing; if the received input is incomplete or ambiguous with respect to the ability to at least partially process the human initiated command, prompting for additional input, by formulating a question to prompt the user, the answer to which is predicted to increase a statistical reliability of a correspondence to the at least one predetermined grammar, wherein the determination of the statistical reliability of the correspondence of the received input to the at least one predetermined grammar comprises a determination of a statistical reliability of a correspondence of the received input to each of a plurality of grammars corresponding to a predetermined plurality of available commands, and a determination of whether a single unambiguous predetermined available is indicated by the received input; and analyze the received input for presence of a human-initiated abort command, and cease the analysis upon recognition of the abort command.
-
-
14. A method for processing speech, comprising:
-
receiving speech as an input; processing the received speech within alternate concurrent instances of a virtual processing environment executing on at least one automated data processing system; automatically communicating, between respective alternate instances of the virtual processing environment, at least one status message to coordinate a reduction in a number of the alternate instances of the virtual processing environment; analyzing; (1) the received speech to determine a command; (2) the received speech to determine data associated with the command; (3) a completeness and unambiguity of the received speech with respect to an ability to execute the command; and if the received speech is complete and unambiguous with respect to the ability to execute the command, executing the command, within at least one virtual processing environment, in conjunction with data associated with the command, and if the received speech is incomplete or ambiguous with respect to the ability to execute the command, prompting for additional input. - View Dependent Claims (15, 16, 17, 18, 19, 20, 21, 22)
-
Specification