Speech interface system and method for control and interaction with applications on a computing system
First Claim
1. A speech processing system, comprising:
- an input port configured to receive a speech input representing at least one of a command and a stream of data by an automated data processing system having a system state;
at least one processor, executing;
a parser configured to distinguish between a command structure and dictation within the speech input;
an analyzer configured to automatically analyze a grammar of the speech input for at least a distinguished command structure, to determine ambiguity and completeness representing a command structure deficiency;
a user feedback system configured to indicate the command structure deficiency and to solicit the user to remedy the deficiency by providing additional speech input, in a contextually appropriate manner, wherein at least the user feedback system is adapted to alter the system state, such that a system state before soliciting the user to remedy the deficiency differs from the system state after soliciting the user to remedy the deficiency; and
an interrupt module, configured to automatically preserve the system state prior to soliciting the user to remedy the deficiency, and after soliciting the user to remedy the deficiency, to automatically restore the preserved prior system state to the system state prior to soliciting the user to remedy the deficiency subject to changes in the system state resulting from execution of the command.
2 Assignments
0 Petitions
Accused Products
Abstract
A speech processing system which exploits statistical modeling and formal logic to receive and process speech input, which may represent data to be received, such as dictation, or commands to be processed by an operating system, application or process. A command dictionary and dynamic grammars are used in processing speech input to identify, disambiguate and extract commands. The logical processing scheme ensures that putative commands are complete and unambiguous before processing. Context sensitivity may be employed to differentiate data and commands. A multi faceted graphic user interface may be provided for interaction with a user to speech enable interaction with applications and processes that do not necessarily have native support for speech input.
24 Citations
20 Claims
-
1. A speech processing system, comprising:
-
an input port configured to receive a speech input representing at least one of a command and a stream of data by an automated data processing system having a system state; at least one processor, executing; a parser configured to distinguish between a command structure and dictation within the speech input; an analyzer configured to automatically analyze a grammar of the speech input for at least a distinguished command structure, to determine ambiguity and completeness representing a command structure deficiency; a user feedback system configured to indicate the command structure deficiency and to solicit the user to remedy the deficiency by providing additional speech input, in a contextually appropriate manner, wherein at least the user feedback system is adapted to alter the system state, such that a system state before soliciting the user to remedy the deficiency differs from the system state after soliciting the user to remedy the deficiency; and an interrupt module, configured to automatically preserve the system state prior to soliciting the user to remedy the deficiency, and after soliciting the user to remedy the deficiency, to automatically restore the preserved prior system state to the system state prior to soliciting the user to remedy the deficiency subject to changes in the system state resulting from execution of the command. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9)
-
-
10. A speech processing method, comprising:
-
determining a variable set of available commands; determining command structures corresponding to the determined set of available commands; processing a natural language speech input representing at least one command with respect to the determined command structures, with at least one automated processor, wherein an interpretation of the natural language speech input is dependent on the determined variable set of available commands; determining and preserving in a system state of the at least one automated processor prior to the natural language speech input; determining if the natural language speech input likely represents a command; determining a completeness and an ambiguity of the represented command with respect to the determined command structures, and if the likely at least one command is too ambiguous or incomplete for execution, obtaining further speech input to reduce ambiguity or incompleteness, wherein at least said obtaining is adapted to change the determined system state, such that a system state before said obtaining further speech input differs from the system state after said obtaining further speech input; executing the represented command with at least one automated processor; and reverting the at least one automated processor to the determined preserved system state with further changes in the system state resulting from executing the represented command by the at least one automated processor. - View Dependent Claims (11, 12, 13, 14, 15, 16)
-
-
17. A speech processing method, comprising:
-
receiving a natural language speech input representing structured commands and unstructured words; analyzing the natural language speech with at least one automated processor for contextual indicia to distinguish between the structured commands, instructing a device at take automated action, and the unstructured words intended as data and not instructing the device to take automated action defined by the unstructured words; determining whether a respective structured command is sufficiently complete to permit at least partial execution, or whether additional command input is required for sufficient completeness to permit at least partial execution; preserving a system state; soliciting further input to increase a completeness of the respective structured command, while causing a first change in the system state, and further determining whether a respective structured command is sufficiently complete to permit at least partial execution; at least partially executing the determined sufficiently complete commands with the at least one automated processor, while causing a second change in the system state; restoring the system state to the preserved system state to reverse the first change in the system state and permit the second change in the system state; and passing the natural language speech containing the unstructured words intended as data to a data sink. - View Dependent Claims (18, 19, 20)
-
Specification