System and method for utilizing speech recognition to efficiently perform data indexing procedures

• US 7,272,562 B2
• Filed: 03/30/2004
• Issued: 09/18/2007
• Est. Priority Date: 03/30/2004
• Status: Active Grant

• Current Assignee: Sony Corporation (Sony Group Corp.), Sony Electronics Inc. (Sony Group Corp.)
• Original Assignee: Sony Corporation (Sony Group Corp.), Sony Electronics Inc. (Sony Group Corp.)
• Inventors: Abrego, Gustavo Hernandez, Koontz, Eugene, Olorenshaw, Lex
• Primary Examiner(s): Hudspeth; David
• Assistant Examiner(s): Jackson; Jakieda R.
• Application Number:
  US10/812,560
• Publication Number:
  US 20050228671A1
• Time in Patent Office: 1,267 Days
• Field of Search: 704/277, 704/235, 704/270
• US Class Current: 704/270
• CPC Class Codes: G06F 16/5866 using information manually generated, e.g. tags, keywords, comments, manually generated location and time information
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence

Abstract

A system and method for utilizing speech recognition to efficiently perform data indexing procedures includes an authoring module that coordinates an authoring procedure for creating an index file that has pattern word sets corresponding to data objects stored in a memory of a host electronic device. The pattern word sets are generated with a speech recognition engine that transforms spoken data descriptions into text data descriptions for creating the pattern word sets. The pattern word sets are associated in the index file with data object identifiers that uniquely identify the corresponding data objects. A retrieval module manages a retrieval procedure in which the speech recognition engine converts a spoken data request into a text data request. The retrieval module compares the text data request and the pattern word sets to identify a requested object identifier for locating a requested data object from among the data objects stored in the memory of the host electronic device.

Claims

1. A system for indexing electronic information, comprising:

an authoring module that coordinates an authoring procedure for creating an index file that includes pattern word sets corresponding to data objects stored in a memory device, said pattern word sets being generated with a speech recognition engine that transforms spoken data descriptions from a system user into text data descriptions, said pattern word sets being associated with data object identifiers that identify said data objects; and

a retrieval module that manages a retrieval procedure in which said speech recognition engine converts a spoken data request into a text data request, said retrieval module comparing said text data request and said pattern word sets to identify a requested object identifier for locating a requested data object from among said data objects stored in said memory device.

2. The system of claim 1 wherein said pattern word sets each include one or more search terms derived from said text data descriptions, said search terms including one or more key words that relate to said data objects.

3. The system of claim 1 wherein an indexing graphical user interface is utilized to create said pattern word sets, said indexing graphical user interface including an object section for displaying representations of said data objects, said indexing graphical user interface also including an editing section for converting said text data descriptions into said pattern word sets.

4. The system of claim 1 wherein said data objects include stored images created by an imaging device, said stored images being saved in said memory device of a host electronic device.

5. The system of claim 4 wherein said authoring module instructs said host electronic device to enter an authoring mode for performing said authoring procedure in response to a verbal authoring-mode command that is recognized by said speech recognition engine.

6. The system of claim 4 wherein said authoring module displays thumbnail representations of said stored images on a display of said host electronic device to facilitate said authoring procedure.

7. The system of claim 4 wherein a selected image is chosen from thumbnail representations of said stored images for creating a corresponding one of said pattern word sets, said selected image being chosen by a system user after viewing said thumbnail representations on an indexing graphical user interface.

8. The system of claim 4 wherein said spoken data description is provided by a system user speaking into a sound sensor of said host electronic device, said spoken data description including a verbal description of a selected image from said stored images.

9. The system of claim 4 wherein said retrieval module instructs said host electronic device to enter a retrieval mode for performing said retrieval procedure in response to a verbal retrieval-mode command that is recognized by said speech recognition engine.

10. The system of claim 4 wherein said spoken data request is provided by a system user speaking into a sound sensor of said host electronic device, said spoken data request including a verbal request related to a selected image from said stored images.

11. The system of claim 4 wherein said retrieval module displays said requested data object on a display of said host electronic device after said retrieval procedure has been performed.

12. The system of claim 1 wherein said speech recognition engine performs speech recognition procedures upon said spoken data descriptions to produce said text data descriptions, said speech recognition engine being implemented as a large-vocabulary continuous speech recognizer that includes acoustic models, a dictionary, and a language model.

13. The system of claim 1 wherein said authoring module displays said text data descriptions on an editing section of an indexing graphical user interface for editing into said pattern word sets.

14. The system of claim 1 wherein said authoring module edits said text data descriptions on an indexing graphical user interface according to editing instructions from a system user to produce said pattern word sets, said editing instructions including a change-word instruction, a delete-word instruction, and an add-word instruction.

15. The system of claim 1 wherein said authoring module stores each of said pattern word sets along with a corresponding one of said data object identifiers into said index file for subsequently performing said retrieval procedure.

16. The system of claim 1 wherein said speech recognition engine performs speech recognition procedures upon said spoken data request to produce said text data request, said speech recognition engine being implemented as a large-vocabulary continuous speech recognizer that includes acoustic models, a dictionary, and a language model.

17. The system of claim 1 wherein said retrieval module accesses said index file for performing a search procedure that identifies said requested data object by seeking search matches between said text data request and said pattern word sets.

18. The system of claim 17 wherein said retrieval module performs said search procedure by utilizing pre-defined priority indicators or pre-defined relationship indicators.

19. The system of claim 18 wherein said pre-defined priority indicators are used to evaluate search terms from said text data request based upon corresponding word lengths.

20. The system of claim 18 wherein said pre-defined relationship indicators are utilized to evaluate search terms from said text data request depending on whether said search terms are located at the beginning or the end of said text data request.

21. A method for indexing electronic information, comprising:

performing an authoring procedure for creating an index file by utilizing an authoring module, said index file including pattern word sets corresponding to data objects stored in a memory device, said pattern word sets being generated with a speech recognition engine that transforms spoken data descriptions from a system user into text data descriptions, said pattern word sets being associated with data object identifiers that identify said data objects; and

utilizing a retrieval module to manage a retrieval procedure in which said speech recognition engine converts a spoken data request into a text data request, said retrieval module comparing said text data request and said pattern word sets to identify a requested object identifier for locating a requested data object from among said data objects stored in said memory device.

22. The method of claim 21 wherein said pattern word sets each include one or more search terms derived from said text data descriptions, said search terms including one or more key words that relate to said data objects.

23. The method of claim 21 wherein an indexing graphical user interface is utilized to create said pattern word sets, said indexing graphical user interface including an object section for displaying representations of said data objects, said indexing graphical user interface also including an editing section for converting said text data descriptions into said pattern word sets.

24. The method of claim 21 wherein said data objects include stored images created by an imaging device, said stored images being saved in said memory device of a host electronic device.

25. The method of claim 24 wherein said authoring module instructs said host electronic device to enter an authoring mode for performing said authoring procedure in response to a verbal authoring-mode command that is recognized by said speech recognition engine.

26. The method of claim 24 wherein said authoring module displays thumbnail representations of said stored images on a display of said host electronic device to facilitate said authoring procedure.

27. The method of claim 24 wherein a selected image is chosen from thumbnail representations of said stored images for creating a corresponding one of said pattern word sets, said selected image being chosen by a system user after viewing said thumbnail representations on an indexing graphical user interface.

28. The method of claim 24 wherein said spoken data description is provided by a system user speaking into a sound sensor of said host electronic device, said spoken data description including a verbal description of a selected image from said stored images.

29. The method of claim 24 wherein said retrieval module instructs said host electronic device to enter a retrieval mode for performing said retrieval procedure in response to a verbal retrieval-mode command that is recognized by said speech recognition engine.

30. The method of claim 24 wherein said spoken data request is provided by a system user speaking into a sound sensor of said host electronic device, said spoken data request including a verbal request related to a selected image from said stored images.

31. The method of claim 24 wherein said retrieval module displays said requested data object on a display of said host electronic device after said retrieval procedure has been performed.

32. The method of claim 21 wherein said speech recognition engine performs speech recognition procedures upon said spoken data descriptions to produce said text data descriptions, said speech recognition engine being implemented as a large-vocabulary continuous speech recognizer that includes acoustic models, a dictionary, and a language model.

33. The method of claim 21 wherein said authoring module displays said text data descriptions on an editing section of an indexing graphical user interface for editing into said pattern word sets.

34. The method of claim 21 wherein said authoring module edits said text data descriptions on an indexing graphical user interface according to editing instructions from a system user to produce said pattern word sets, said editing instructions including a change-word instruction, a delete-word instruction, and an add-word instruction.

35. The method of claim 21 wherein said authoring module stores each of said pattern word sets along with a corresponding one of said data object identifiers into said index file for subsequently performing said retrieval procedure.

36. The method of claim 21 wherein said speech recognition engine performs speech recognition procedures upon said spoken data request to produce said text data request, said speech recognition engine being implemented as a large-vocabulary continuous speech recognizer that includes acoustic models, a dictionary, and a language model.

37. The method of claim 21 wherein said retrieval module accesses said index file for performing a search procedure that identifies said requested data object by seeking search matches between said text data request and said pattern word sets.

38. The method of claim 37 wherein said retrieval module performs said search procedure by utilizing pre-defined priority indicators and pre-defined relationship indicators.

39. The method of claim 38 wherein said pre-defined priority indicators are used to evaluate search terms from said text data request based upon corresponding word lengths.

40. The method of claim 38 wherein said pre-defined relationship indicators are utilized to evaluate search terms from said text data request depending on whether said search terms are located at the beginning or the end of said text data request.

41. The method of claim 37 wherein said retrieval module performs said search procedure by utilizing pre-defined priority indicators to evaluate search terms from said text data request.

42. The method of claim 37 wherein said retrieval module performs said search procedure by utilizing pre-defined relationship indicators to evaluate search terms from said text data request.

43. A computer-readable medium comprising program instructions for indexing electronic information by:

performing an authoring procedure for creating an index file by utilizing an authoring module, said index file including pattern word sets corresponding to data objects stored in a memory device, said pattern word sets being generated with a speech recognition engine that transforms spoken data descriptions from a system user into text data descriptions for creating said pattern word sets, said pattern word sets being associated with data object identifiers that identify said data objects; and

utilizing a retrieval module to manage a retrieval procedure in which said speech recognition engine converts a spoken data request into a text data request, said retrieval module comparing said text data request and said pattern word sets to identify a requested object identifier for locating a requested data object from among said data objects stored in said memory device.

44. A system for cataloguing electronic information, comprising:

means for performing an authoring procedure for creating an index file that includes pattern word sets corresponding to data objects stored in a memory device, said pattern word sets being generated with a speech recognition engine that transforms spoken data descriptions from a system user into text data descriptions for creating said pattern word sets, said pattern word sets being associated with data object identifiers that identify said data objects; and

means for managing a retrieval procedure in which said speech recognition engine converts a spoken data request into a text data request, said means for managing then comparing said text data request and said pattern word sets to identify a requested object identifier for locating a requested data object from among said data objects stored in said memory device.