Assigning meanings to utterances in a speech recognition system

• US 20040162729A1
• Filed: 02/18/2004
• Published: 08/19/2004
• Est. Priority Date: 12/31/1992
• Status: Active Grant

• Current Assignee: Apple Inc.
• Original Assignee: Apple Inc.
• Inventors: Strong, Robert Don
• Granted Patent: US 7,127,394 B2
• Time in Patent Office: Days
• US Class Current: 704/259
• CPC Class Codes: G10L 15/1822 Parsing for meaning understanding
• CPC Class Codes: G10L 2015/228 of application context

Abstract

Assigning meanings to spoken utterances in a speech recognition system. A plurality of speech rules is generated, each of the of speech rules comprising a language model and an expression associated with the language model. At one interval (e.g. upon the detection of speech in the system), a current language model is generated from each language model in the speech rules for use by a recognizer. When a sequence of words is received from the recognizer, a set of speech rules which match the sequence of words received from the recognizer is determined. Each expression associated with the language model in each of the set of speech rules is evaluated, and actions are performed in the system according to the expressions associated with each language model in the set of speech rules.

Claims

1. A method of associating meanings to utterances in a speech recognition system comprising the following steps:

a. generating a plurality of speech rules, each of said plurality of speech rules comprising a language model and an expression associated with said language model;

b. upon the detection of speech in said speech recognition system, generating a current language model from each said language model in said plurality of speech rules for use by a recognizer;

c. receiving a sequence of words from said recognizer, and determining a set of said speech rules which match said sequence of words received from said recognizer; and

d. evaluating said expression associated with said language model in each speech rule of said set of speech rules, and performing actions in said system according to said expressions associated with each said language model in said set of speech rules.

2. The method of claim 1 wherein each said expression comprises an arithmetic calculation for said variables in said speech recognition system.

3. The method of claim 1 wherein each said language model in each of said plurality of speech rules comprises recursive references to other language models.

4. The method of claim 3 wherein each of said other language models comprise expressions associated with each of said other language models.

5. The method of claim 4 wherein said evaluation step comprises the steps of evaluating each of said expressions for said other language models prior to evaluating each said language model in each speech rule of said set of speech rules.

6. An apparatus for associating meanings to utterances in a speech recognition system comprising the following steps:

a. means for generating a plurality of speech rules, each of said plurality of speech rules comprising a language model and an expression associated with said language model;

b. means for generating a current language model from each said language model in said plurality of speech rules for use by a recognizer upon the detection of speech in said speech recognition system;

c. means for receiving a sequence of words from said recognizer, and determining a set of said speech rules which match said sequence of words received from said recognizer; and

d. means for evaluating said expression associated with said language model in each speech rule of said set of speech rules, and performing actions in said system according to said expressions associated with each said language model in said set of speech rules.