Claims
- 1. A method of recognizing punctuation in computer-implemented speech recognition, the method comprising:
performing speech recognition on an utterance to produce a recognition result for the utterance; identifying a non-verbalized punctuation mark in a recognition result; and formatting the recognition result based on the identification.
- 2. The method as in claim 1 wherein identifying the non-verbalized punctuation mark includes predicting the non-verbalized punctuation mark using at least one text feature and at least one acoustic feature related to the utterance.
- 3. The method as in claim 2 wherein the acoustic feature includes a period of silence.
- 4. The method as in claim 2 wherein the acoustic feature includes a function of pitch of words near the period of silence.
- 5. The method as in claim 2 wherein the acoustic feature includes an average pitch of words near the period of silence.
- 6. The method as in claim 2 wherein the acoustic feature includes a ratio of an average pitch of words near the period of silence.
- 7. The method as in claim 1 wherein formatting the recognition result includes controlling or altering spacing relative to the non-verbalized punctuation mark.
- 8. The method as in claim 1 wherein formatting the recognition result includes controlling or altering capitalization of words relative to the non-verbalized punctuation mark.
- 9. The method as in claim 1 wherein:
the non-verbalized punctuation mark includes a period, and formatting the recognition result includes inserting an extra space after the period and capitalizing a next word following the period.
- 10. The method as in claim 1 further comprising:
selecting a portion of the recognition result to be corrected that includes the non-verbalized punctuation mark; and correcting the portion of the recognition result that includes the non-verbalized punctuation mark with one of a number of correction choices.
- 11. The method as in claim 10 wherein at least one of the correction choices includes a change to the non-verbalized punctuation mark.
- 12. The method as in claim 10 wherein at least one of the correction choices does not include the non-verbalized punctuation mark.
- 13. An apparatus comprising a computer-readable medium having instructions stored thereon that when executed by a machine result in at least the following:
performing speech recognition on an utterance to produce a recognition result for the utterance; identifying a non-verbalized punctuation mark in a recognition result; and formatting the recognition result based on the identification.
- 14. A method of correcting incorrect text associated with recognition errors in computer-implemented speech recognition, comprising:
performing speech recognition on an utterance to produce a recognition result for the utterance, wherein the recognition result includes non-verbalized punctuation; selecting a portion of the recognition result to be corrected that includes the non-verbalized punctuation; and correcting the portion of the recognition result that includes the non-verbalized punctuation with one of a number of correction choices.
- 15. The method as in claim 14 wherein at least one of the correction choices includes a change to the non-verbalized punctuation.
- 16. The method as in claim 14 wherein at least one of the correction choices does not include the non-verbalized punctuation.
- 17. The method as in claim 14 where the non-verbalized punctuation includes a non-verbalized punctuation mark.
- 18. The method as in claim 14 wherein correcting the portion of the recognition result includes changing the non-verbalized punctuation and reformatting text surrounding the non-verbalized punctuation to be grammatically consistent with the changed non-verbalized punctuation.
- 19. The method as in claim 18 wherein changing the non-verbalized punctuation and reformatting text surrounding the non-verbalized punctuation is in response to a single user action.
- 20. An apparatus comprising a computer-readable medium having instructions stored thereon that when executed by a machine result in at least the following:
performing speech recognition on an utterance to produce a recognition result for the utterance, wherein the recognition result includes non-verbalized punctuation; selecting a portion of the recognition result to be corrected that includes the non-verbalized punctuation; and correcting the portion of the recognition result that includes the non-verbalized punctuation with one of a number of correction choices.
- 21. A method of recognizing punctuation in computer-implemented speech recognition dictation, the method comprising:
performing speech recognition on an utterance to produce a recognition result for the utterance; identifying a non-verbalized punctuation mark in a recognition result; and determining where to insert the non-verbalized punctuation mark within the recognition result based on the identification using at least one text feature and at least one acoustic feature related to the utterance to predict where to insert the non-verbalized punctuation mark.
- 22. The method as in claim 21 wherein the acoustic feature includes a period of silence.
- 23. The method as in claim 21 wherein the acoustic feature includes a function of pitch of words near the period of silence.
- 24. The method as in claim 21 wherein the acoustic feature includes an average pitch of words near the period of silence.
- 25. The method as in claim 21 wherein the acoustic feature includes a ratio of an average pitch of words near the period of silence.
- 26. An apparatus comprising a computer-readable medium having instructions stored thereon that when executed by a machine result in at least the following:
performing speech recognition on an utterance to produce a recognition result for the utterance; identifying a non-verbalized punctuation mark in a recognition result; and determining where to insert the non-verbalized punctuation mark within the recognition result based on the identification using at least one text feature and at least one acoustic feature related to the utterance to predict where to insert the non-verbalized punctuation mark.
- 27. A graphical user interface for correcting incorrect text associated with recognition errors in computer-implemented speech recognition, comprising:
a window to display a selected recognition result including non-verbalized punctuation associated with an utterance; and a list of recognition alternatives with at least one of the recognition alternatives including a change to the non-verbalized punctuation and associated adjustments in spacing and other punctuation.
- 28. The graphical user interface of claim 27 wherein the non-verbalized punctuation includes a period.
- 29. The graphical user interface of claim 27 wherein the non-verbalized punctuation includes a comma.
- 30. The graphical user interface of claim 27 wherein:
the change to the non-verbalized punctuation includes a change from a period to a comma, and the associated adjustments in spacing and other punctuation includes removing a space after the comma and uncapitalizing a word following the comma.
- 31. The graphical user interface of claim 27 wherein:
the change to the non-verbalized punctuation includes a change from a comma to a period, and the associated adjustments in spacing and other punctuation includes adding a space after the period and capitalizing a word following the period.
CROSS REFERENCE TO RELATED APPLICATIONS
[0001] This application claims the benefit of U.S. Provisional Application No. 60/428,232, filed Nov. 22, 2002, and titled “Automatic Insertion of Nonverbalized Punctuation In Speech Recognition,” which is incorporated by reference in its entirety.
Provisional Applications (1)
|
Number |
Date |
Country |
|
60428232 |
Nov 2002 |
US |