Disambiguation systems and methods for use in generating grammars
View Patent ↗A method and system for addressing disambiguation issues in interactive applications by creating a disambiguation system for generating complex grammars that includes homonym detection and grouping, and provides optimization feedback that eliminates time-consuming and repetitive iterative steps during the grammar generation portion of the interactive application configuration.
1. A method for generating grammars for use in an interactive application, the method comprising:
providing data records to be used in a grammar database, each data record including one or more fields;
prior to identifying, in response to a user-issued query, a set of the data records that satisfy the query and on which disambiguation is to be performed, receiving human input specifying on which of the fields of each of the data records homonym and/or synonym disambiguation is to be performed;
applying a disambiguation engine to each data record to generate disambiguated data, wherein the disambiguation engine resolves ambiguities relating to at least one homophone and/or at least one synonym in the data in the specified fields; and
based on the disambiguated data, generating one or more grammar files for use in the interactive application.
2. The method of claim 1 , further comprising providing an end-user interactive application that selects appropriate grammars from the one or more grammar files upon receipt of user responses to system prompts.
3. The method of claim 1 , further comprising displaying, on a display, ambiguities resolved by the disambiguation engine.
4. The method of claim 1 , further comprising identifying annotation fields, the annotation fields representing additional disambiguation fields used to further resolve ambiguities.
5. The method of claim 1 , further comprising calculating grammar generation statistics in order to optimize future grammar file generation.
6. The method of claim 5 , wherein upon determining that a the number of grammar files generated exceeds a threshold value, further comprising suggesting the use of additional disambiguation fields.
7. The method of claim 1 , wherein the one or more grammar files are generated for use in an interactive voice recognition system.
8. The method of claim 1 , wherein the disambiguation engine resolves ambiguities relating to at least one homophone in the data in the specified fields.
9. A system for generating grammars for use in an interactive application, the system comprising:
a data source, stored on at least one computer-readable storage medium, the data source comprising data records to be used in a grammar database, each data record including one or more fields;
an input for, prior to identifying, in response to a user-issued query, a set of the data records that satisfy the query and on which disambiguation is to be performed, receiving human input that specifies on which of the fields of each of the data records homonym and/or synonym disambiguation is to be performed;
a disambiguation engine stored on the at least one computer-readable storage medium, wherein the disambiguation engine resolves ambiguities relating to at least one homophone and/or at least one synonym in data in the specified fields of the one or more disambiguation fields to generate disambiguated data; and
a grammar generation engine stored on the at least one computer-readable storage medium, wherein the grammar generation engine generates one or more grammar files based on the disambiguated data for use in the interactive application.
10. The system of claim 9 , further comprising an application generation engine for providing an end-user interactive application that selects appropriate grammars from the one or more grammar files upon receipt of user responses to system prompts.
11. The system of claim 9 , further comprising display means for displaying the results of the one or more grammar files generated by the grammar generation engine.
12. The system of claim 11 , wherein the display means displays annotation fields representing additional disambiguation fields used to further resolve ambiguities.
13. The system of claim 9 , further comprising a results collection module for generating statistics in order to optimize future grammar generation.
14. The system of claim 13 , wherein the results collection module suggests additional disambiguation fields for identification upon determining that a number of grammar files generated is too large.
15. The system of claim 9 , wherein the one or more grammar files are generated for use in an interactive voice recognition system.
16. At least one tangible computer readable storage medium encoded with instructions that, when executed by at least one processing unit, perform a method for generating grammars for use in an interactive application, wherein the grammars are generated from a data record source and each data record includes one or more fields, the method comprising:
prior to identifying, in response to a user-issued query, a set of the data records that satisfy the query and on which disambiguation is to be performed, receiving human input specifying on which of the fields of each of the data records homonym and/or synonym disambiguation is to be performed;
resolving ambiguities relating to at least one homophone and/or at least one synonym in data in the specified fields to generate disambiguated data; and
based on the disambiguated data, generating one or more grammar files for use in the interactive application.
17. The at least one tangible computer readable storage medium of claim 16 , the method further comprising:
providing an end-user interactive application that selects appropriate grammars from the one or more grammar files upon receipt of user responses to system prompts.
18. The at least one tangible computer readable storage medium of claim 16 , the method further comprising:
identifying annotation fields, the annotation fields representing additional disambiguation fields used to further resolve ambiguities.
19. The at least one tangible computer readable storage medium of claim 16 , the method further comprising:
calculating grammar generation statistics in order to provide optimization advice for future grammar file generation.
20. The at least one tangible computer readable storage medium of claim 19 , wherein the optimization advice is based on voice recognition engine capabilities.
21. The at least one tangible computer readable storage medium of claim 19 , the method further comprising:
upon determining that a number of grammar files generated exceeds a threshold value, suggesting the use of additional disambiguation fields.
22. The at least one tangible computer readable storage medium of claim 16 , wherein the one or more grammar files are generated for use in an interactive voice recognition system.