Command parsing and rewrite system
First Claim
1. An enhanced speech recognition system having dictation capability and the capability of converting an unprompted user-generated word phrase into a command string for causing a computer application to perform actions as directed by said word phrase comprising:
- a speech recognition engine providing a dictation output and candidate word phrases to a second output;
a command interpreter, coupled to the second output, the command interpreter having;
(i) a parser for completely parsing a candidate word phrase into a parse tree;
(ii) a plurality of predefined rewrite rules divided and grouped into a plurality of phases; and
(iii) a rewriting component, for applying at least one of said predefined rewrite rules to said parse tree, for rewriting said parse tree according to said at least one predefined rewrite rule so as to produce said command string, wherein all predefined rewrite rules grouped into one of said plurality of phases are applied to said parse tree before applying predefined rewrite rules grouped into another of said plurality of phases to said parse tree.
9 Assignments
0 Petitions
Accused Products
Abstract
A system and method of allowing a user to control a computer application with spoken commands, include the steps of processing the spoken commands with a Speech Recognition application into candidate word phrases, and parsing at least one candidate word phrase with a Context Free Grammar (CFG) parser, into a parse tree. A plurality of predefined rewrite rules grouped into a plurality of phases applied are to the parse tree, for rewriting the parse tree. Each of the plurality of rewrite rules includes a pattern matching portion, for matching at least a part of the parse tree, and a rewrite component, for rewriting the matched part. A command string is produced by traversing each terminal node of the modified parse tree. The command string is sent to an interpreter application or directly to the computer application. Possible applications include word processing and other voice-entry systems.
-
Citations
23 Claims
-
1. An enhanced speech recognition system having dictation capability and the capability of converting an unprompted user-generated word phrase into a command string for causing a computer application to perform actions as directed by said word phrase comprising:
-
a speech recognition engine providing a dictation output and candidate word phrases to a second output; a command interpreter, coupled to the second output, the command interpreter having; (i) a parser for completely parsing a candidate word phrase into a parse tree; (ii) a plurality of predefined rewrite rules divided and grouped into a plurality of phases; and (iii) a rewriting component, for applying at least one of said predefined rewrite rules to said parse tree, for rewriting said parse tree according to said at least one predefined rewrite rule so as to produce said command string, wherein all predefined rewrite rules grouped into one of said plurality of phases are applied to said parse tree before applying predefined rewrite rules grouped into another of said plurality of phases to said parse tree. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11)
-
-
12. A method of an enhanced speech recognition system for interpreting an unprompted user-generated word phrase and directing a computer application to perform actions according to said word phrase, comprising the steps of:
-
providing a first output having recognized dictated text, and a separate second output having a plurality of candidate word phrases, the first and second outputs being mutually exclusive; parsing said word phrase from the candidate word phrase output into a complete parse tree representing said candidate word phrase; applying a plurality of predefined rewrite rules to said parse tree, for rewriting said parse tree into at least one modified parse tree, the plurality of predefined rewrite rules being divided and grouped into a plurality of phases each of said plurality of predefined rewrite rules including a pattern matching portion, for comparison to and matching at least a part of said parse tree, and also including a rewriting portion, for rewriting a matched part of said parse tree, each of the predefined rewrite rules grouped in one of the plurality of phases being applied before applying predefined rewrite rules grouped into another of the plurality of phases; producing a command string by traversing nodes of said at least one modified parse tree; and providing said command string to said computer application. - View Dependent Claims (13, 14, 15, 16, 17, 18, 19, 20, 21, 22)
-
-
23. A method of an enhanced speech recognition system for allowing a user to control a computer application with spoken commands, comprising the steps of:
-
converting an unprompted user-generated word phrase into electrical signals representing said word phrase; processing said electrical signals, with a Speech Recognition application that produces a first output having recognized dictated text, and a separate second output having a plurality of candidate word phrases, the first and second outputs being mutually exclusive; parsing at least one candidate word phrase with a Context Free Grammar (CFG) parser, into a complete parse tree; applying a plurality of predefined rewrite rules grouped into a plurality of phases to said parse tree, for rewriting said parse tree into at least one modified parse tree, each of said plurality of predefined rewrite rules including a pattern matching portion, for matching at least a part of said parse tree, and also including a rewriting portion, for rewriting said matched part of said parse tree, the predefined rewrite rules grouped in one of the plurality of phases being applied before applying predefined rewrite rules in another of the plurality of phases; producing a command string by traversing nodes of said at least one modified parse tree; providing said command string to an interpreter application; and directing said interpreter application to execute said command string, for causing said interpreter application to instruct said computer application to perform actions appropriate for said spoken command.
-
Specification