Speech recognition apparatus and its method and program

• US 20050086057A1
• Filed: 11/13/2002
• Published: 04/21/2005
• Est. Priority Date: 11/22/2001
• Status: Abandoned Application

• Current Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Original Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Inventors: Sakai, Keiichi, Kosaka, Tetsuo, Yamamoto, Hiroki
• Application Number:
  US10/490,696
• Publication Number:
  US 20050086057A1
• Time in Patent Office: Days
• US Class Current: 704/251
• CPC Class Codes: G10L 15/19 Grammatical context, e.g. disambiguation of the recognition hypotheses based on word sequence rules
• CPC Class Codes: G10L 15/193 Formal grammars, e.g. finite state automata, context free grammars or word networks
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 2015/228 of application context

Abstract

Data including a description for displaying input columns and a description concerning speech recognition grammars for the input columns is read. By using speech recognition grammars corresponding to a plurality of input columns displayed on the basis of the data, speech recognition on the input speech is conducted. On the basis of the speech recognition grammars, input columns of input destinations of the speech recognition result are determined from a plurality of input columns and the input columns of input destinations are displayed in corresponding input columns.

Claims

1. A speech recognition apparatus for recognizing input speech, comprising:

reading means for reading hypertext document data, the hypertext document data including a description for displaying input columns and a description concerning speech recognition grammar data to be applied to input speech for said input columns;

speech recognition means for conducting speech recognition on said input speech by using speech recognition grammar data corresponding to a plurality of input columns displayed on the basis of said hypertext document data; and

display means for determining input columns of input destinations of word groups obtained from a speech recognition result of said speech recognition means from said plurality of input columns on the basis of said speech recognition grammar data, and displaying said input columns of input destinations in corresponding input columns.

2. The speech recognition apparatus according to claim 1, further comprising:

specification means for specifying a plurality of input columns displayed on the basis of said hypertext document data,

wherein said speech recognition means conducts speech recognition on said input speech by using speech recognition grammar data corresponding to said plurality of input columns specified by said specification means.

3. The speech recognition apparatus according to claim 2, wherein said specification means can specify said plurality of input columns simultaneously.

4. The speech recognition apparatus according to claim 1, wherein said display means determines input columns of input destinations of word groups obtained from a speech recognition result of said speech recognition means from said plurality of input columns on the basis of said speech recognition grammar data, and displaying said input columns of input destinations in corresponding input columns simultaneously.

5. The speech recognition apparatus according to claim 1, wherein said hypertext document data and said speech recognition grammar data are managed in an external terminal connected to said speech recognition apparatus via a network.

6. The speech recognition apparatus according to claim 1, further comprising:

analysis means for analyzing said hypertext document data;

first holding means for acquiring said speech recognition grammar data corresponding to said input columns from an analysis result of said analysis means and holding said speech recognition grammar data so as to be associated with said input columns; and

second holding means for holding words so as to be associated with said input columns.

7. The speech recognition apparatus according to claim 1, further comprising:

morphological analysis means for effecting a morphological analysis on said speech recognition result,

wherein said display means determines input columns of input destinations of word groups obtained from said speech recognition result from said plurality of input columns on the basis of said speech recognition grammar data and a result of said morphological analysis of said morphological analysis means on said speech recognition result of said speech recognition means, and displaying said input columns of input destinations in corresponding input columns.

8. A speech recognition apparatus for recognizing input speech, comprising:

reading means for reading hypertext document data, the hypertext document data including a description for displaying input columns and a description concerning speech recognition grammar data to be applied to input speech for said input columns;

analysis means for analyzing said hypertext document;

generation means for generating speech recognition grammar data corresponding to predetermined input columns including a plurality of input columns included in said hypertext document on the basis of the analysis result of said analysis means;

speech recognition means for conducting speech recognition on said input speech by using speech recognition grammar data corresponding to said predetermined input columns displayed on the basis of said hypertext document data; and

display means for determining input columns of input destinations of word groups obtained from a speech recognition result of said speech recognition means from a plurality of input columns forming said predetermined input columns on the basis of said speech recognition grammar data, and displaying said input columns of input destinations in corresponding input columns.

9. The speech recognition apparatus according to claim 8, wherein

said analysis means comprises extraction means for extracting a description having no corresponding speech recognition grammar data, from among descriptions for displaying input columns included in said hypertext document, and

said generation means generates speech recognition grammar data corresponding to input columns corresponding to said description, on the basis of said description extracted by said extraction means.

10. The speech recognition apparatus according to claim 8, wherein

said analysis means comprises extraction means that extracts a predetermined description for generating speech recognition grammar data included in said hypertext document, and

said generation means generates speech recognition grammar data corresponding to said predetermined input column, on the basis of speech recognition grammar data specified on the basis of the predetermined description extracted by said extraction means.

11. The speech recognition apparatus according to claim 8, wherein

said generation means comprises extraction means that extracts a description for making text data of a display subject included in said hypertext document a speech recognition subject, and

said generation means generates speech recognition grammar data including said text data corresponding to input columns that corresponds to said description, on the basis of the description extracted by said extraction step.

12. A speech recognition method for recognizing input speech, comprising:

a reading step of reading hypertext document data, the hypertext document data including a description for displaying input columns and a description concerning speech recognition grammar data to be applied to input speech for said input columns;

a speech recognition step of conducting speech recognition on said input speech by using speech recognition grammar data corresponding to a plurality of input columns displayed on the basis of said hypertext document data; and

a display step of determining input columns of input destinations of word groups obtained from a speech recognition result of said speech recognition step from said plurality of input columns on the basis of said speech recognition grammar data, and displaying said input columns of input destinations in corresponding input columns.

13. The speech recognition method according to claim 12, further comprising:

a specification step of specifying a plurality of input columns displayed on the basis of said hypertext document data,

wherein said speech recognition step conducts speech recognition on said input speech by using speech recognition grammar data corresponding to said plurality of input columns specified by said specification step.

14. The speech recognition method according to claim 13, wherein said specification step can specify said plurality of input columns simultaneously.

15. The speech recognition method according to claim 12, wherein said display step determines input columns of input destinations of word groups obtained from a speech recognition result of said speech recognition step from said plurality of input columns on the basis of said speech recognition grammar data, and displaying said input columns of input destinations in corresponding input columns simultaneously.

16. The speech recognition method according to claim 12, wherein said hypertext document data and said speech recognition grammar data are managed in an external terminal connected to a pertinent speech recognition apparatus via a network.

17. The speech recognition method according to claim 12, further comprising:

an analysis step of analyzing said hypertext document data;

a first holding step of acquiring said speech recognition grammar data corresponding to said input columns from an analysis result of said analysis step and holding said speech recognition grammar data so as to be associated with said input columns; and

a second holding step of holding words so as to be associated with said input columns.

18. The speech recognition method according to claim 12, further comprising:

a morphological analysis step of effecting a morphological analysis on said speech recognition result,

wherein said display step determines input columns of input destinations of word groups obtained from said speech recognition result from said plurality of input columns on the basis of said speech recognition grammar data and a result of said morphological analysis of said morphological analysis step on said speech recognition result of said speech recognition step, and displaying said input columns of input destinations in corresponding input columns.

19. A speech recognition method for recognizing input speech, comprising:

a reading step of reading hypertext document data, the hypertext document data including a description for displaying input columns and a description concerning speech recognition grammar data to be applied to input speech for said input columns;

an analysis step of analyzing said hypertext document;

a generation step of generating speech recognition grammar data corresponding to predetermined input columns including a plurality of input columns included in said hypertext document;

a speech recognition step of conducting speech recognition on said input speech by using speech recognition grammar data corresponding to said predetermined input columns displayed on the basis of said hypertext document data; and

a display step of determining input columns of input destinations of word groups obtained from a speech recognition result of said speech recognition step from a plurality of input columns forming said predetermined input columns on the basis of said speech recognition grammar data, and displaying said input columns of input destinations in corresponding input columns.

20. The speech recognition method according to claim 19, wherein

said analysis step comprises an extraction step of extracting a description having no corresponding speech recognition grammar data, from among descriptions for displaying input columns included in said hypertext document, and

said generation step generates speech recognition grammar data corresponding to input columns corresponding to said description, on the basis of said description extracted by said extraction step.

21. The speech recognition method according to claim 19, wherein

said analysis step comprises extraction step that extracts a predetermined description for generating speech recognition grammar data included in said hypertext document, and

said generation step generates speech recognition grammar data corresponding to said predetermined input column, on the basis of speech recognition grammar data specified on the basis of the predetermined description extracted by said extraction step.

22. The speech recognition method according to claim 19, wherein

said analysis step comprises extraction step that extracts a description for making text data of a display subject included in said hypertext document a speech recognition subject, and

said generation step generates speech recognition grammar data including said text data corresponding to input columns that corresponds to said description, on the basis of the description extracted by said extraction step.

23. A program for causing a computer to function to conduct speech recognition for recognizing input speech, comprising:

a program code for a reading step of reading hypertext document data, the hypertext document data including a description for displaying input columns and a description concerning speech recognition grammar data to be applied to input speech for said input columns;

a program code for a speech recognition step of conducting speech recognition on said input speech by using speech recognition grammar data corresponding to a plurality of input columns displayed on the basis of said hypertext document data; and

a program code for a display step of determining input columns of input destinations of word groups obtained from a speech recognition result of said speech recognition step from said plurality of input columns on the basis of said speech recognition grammar data, and displaying said input columns of input destinations in corresponding input columns.

24. A program for causing a computer to function to conduct speech recognition for recognizing input speech, comprising:

a program code for a reading step of reading hypertext document data, the hypertext document data including a description for displaying input columns and a description concerning speech recognition grammar data to be applied to input speech for said input columns;

a program code for an analysis step of analyzing said hypertext document;

a program code for a generation step of generating speech recognition grammar data corresponding to predetermined input columns including a plurality of input columns included in said hypertext document on the basis of the analysis result of said analysis, step;

a program code for a speech recognition step of conducting speech recognition on said input speech by using speech recognition grammar data corresponding to said predetermined input columns displayed on the basis of said hypertext document data; and

a program code for a display step of determining input columns of input destinations of word groups obtained from a speech recognition result of said speech recognition step from a plurality of input columns forming said predetermined input columns on the basis of said speech recognition grammar data, and displaying said input columns of input destinations in corresponding input columns.

25. A speech recognition apparatus comprising:

reading means for reading data, the data including a description for displaying input columns and a description concerning speech recognition grammars for said input columns;

speech recognition means for conducting speech recognition on said input speech by using said speech recognition grammars; and

display means for determining input columns of input destinations of said speech recognition result from said plurality of input columns on the basis of said speech recognition grammars, and displaying said input columns of input destinations in corresponding input columns.

26. A speech recognition method comprising:

a reading step of reading data, the data including a description for displaying input columns and a description concerning speech recognition grammars for said input columns;

a speech recognition step of conducting speech recognition on said input speech by using said speech recognition grammars; and

a display step of determining input columns of input destinations of said speech recognition result from said plurality of input columns on the basis of said speech recognition grammars, and displaying said input columns of input destinations in corresponding input columns.

27. A program for causing a computer to function to conduct speech recognition for recognizing input speech, comprising:

a program code for a reading step of reading data, the data including a description for displaying input columns and a description concerning speech recognition grammars for said input columns;

a program code for a speech recognition step of conducting speech recognition on said input speech by using said speech recognition grammars; and

a program code for a display step of determining input columns of input destinations of said speech recognition result from said plurality of input columns on the basis of said speech recognition grammars, and displaying said input columns of input destinations in corresponding input columns.