Generic natural language service creation environment

• US 6,684,183 B1
• Filed: 12/06/1999
• Issued: 01/27/2004
• Est. Priority Date: 12/06/1999
• Status: Expired due to Fees

• Current Assignee: Comverse Limited (Mavenir Group Holdings Ltd.)
• Original Assignee: Comverse Limited (Mavenir Group Holdings Ltd.)
• Inventors: Peʼer, Eran, Deddi, Dorit, Firt, Erez, Korall, Toby
• Primary Examiner(s): EDOUARD, PATRICK NESTOR
• Application Number:
  US09/455,155
• Time in Patent Office: 1,513 Days
• Field of Search: 704/8, 704/9, 704/10, 704/1, 704/255, 704/257, 704/231, 704/251
• US Class Current: 704/9
• CPC Class Codes: G06F 40/211 Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars
• CPC Class Codes: G06F 40/253 Grammatical analysis; Style critique
• CPC Class Codes: G10L 15/193 Formal grammars, e.g. finite state automata, context free grammars or word networks
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence

Abstract

A system that allows a user to define a set of grammar rules containing variable types and terms associated with those types. The user enters sample phrases and identifies terms in the sample phrases that he wishes to define as a type, and then specifies which type that term corresponds to. The system, using the stored terms and types, generates grammar files which are specific to a particular automatic speech recognition engine.

Claims

1. A method of supplementing a grammar, comprising:

defining types;

entering example phrases on a computer, the example phrases comprising terms not used in the grammar;

identifying, in the example phrases, the terms not in the grammar and their respective application grammatical type by sending the phrases to a parser; and

storing the identified terms and their respective types in the grammar to supplement the grammar with the terms not used in the grammar.

2. The method recited in claim 1, further comprising automatically generating a grammar file readable by a specified automatic speech recognition engine.

3. The method recited in claim 1, further comprising reading a grammar file dedicated for a specific automatic speech recognition engine.

4. The method recited in claim 1, further comprising:

receiving a query from a user;

comparing portions of the query with the terms, wherein when a term and a portion of the query matches, the term is associated with the term'"'"'s respective type; and

outputting the terms that have associated with them a type, and outputting the associated type.

5. The method recited in claim 4, wherein the query is spoken over a telephone.

6. The method recited in claim 4, further comprising:

checking if one or more required type(s) does not have an association; and

unless the checking determines all the required types have an association, asking the user to specify a term for each undefined type.

7. A computer readable storage storing a program instructing a computer to perform a method of supplementing a grammar, comprising:

allowing a user to define types;

allowing the user to enter example phrases on a computer, the example phrases comprising terms not used in a grammar;

allowing the user to identify, in the example phrases, terms and their respective application grammatical type based on the users knowledge of the grammar; and

storing the identified terms and their respective types in the grammar to supplement the grammar with the terms not used in the grammar.

8. The computer readable storage medium as recited in claim 7, further comprising generating a grammar file readable by a specified automatic speech recognition engine.

9. The computer readable storage medium as recited in claim 7, further comprising reading a grammar file dedicated for a specific automatic speech recognition engine.

10. The computer readable storage medium recited in claim 7, further comprising:

receiving a query from a user;

comparing portions of the query with the terms, wherein when a term and a portion of the query matches, the term is associated with the term'"'"'s respective type; and

outputting the terms that have associated with them a type, and outputting the associated type.

11. The computer readable storage medium recited in claim 10, wherein the query is spoken over a telephone.

12. The computer readable storage medium recited in claim 10, further comprising:

checking if one or more required type(s) does not have an association; and

unless the checking determines all the required types have an association, asking the user to specify a term for each unassociated type.

13. A method comprising:

defining types;

entering example phrases on a computer;

identifying, in the example phrases, terms and their respective type;

storing the identified terms and their respective types;

automatically generating a grammar file readable by a specified automatic speech recognition engine;

reading a grammar file dedicated for a specific automatic speech recognition engine;

receiving a query on the computer via a telephone;

comparing, on the computer, portions of the query with the terms, wherein when a term and a portion of the query matches, the term is associated with the term'"'"'s respective type;

outputting the terms that have associated with them a type, and outputting the associated type;

checking if one or more required type(s) does not have an association;

unless the checking determines all the required types have an association, asking the user to specify a term for each undefined type;

providing a type window allowing a user to enter a plurality of types;

providing an example window allowing the user to enter example phrases containing words;

receiving a highlighted selection of word(s) from the example phrases in the example window;

providing a type selection window displaying the entered types;

receiving a selected type from the type selection window;

removing the type selection window; and

storing the selected word(s) and the selected type in a storage device.

14. A method of supplementing a grammar using example phrases, comprising:

identifying, in the example phrases, terms not in the grammar and identifying their respective application grammatical type using a parser; and

storing the identified terms and their respective types in the grammar to supplement the grammar with the terms not used in the grammar.