Word recognition using word transformation commands

• US 7,634,403 B2
• Filed: 09/24/2004
• Issued: 12/15/2009
• Est. Priority Date: 09/05/2001
• Status: Active Grant

• Current Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Original Assignee: Voice Signal Technologies Incorporated (Microsoft Corporation)
• Inventors: Roth, Daniel L., Johnston, David F., Cohen, Jordan R.
• Primary Examiner(s): MCFADDEN, SUSAN IRIS
• Application Number:
  US10/949,974
• Publication Number:
  US 20050038653A1
• Time in Patent Office: 1,908 Days
• Field of Search: 704/270, 704/251, 704/235
• US Class Current: 704/235
• CPC Class Codes: G10L 15/19 Grammatical context, e.g. disambiguation of the recognition hypotheses based on word sequence rules
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue

Abstract

Word recognition enables a user to have a selected transformation performed on a given word produced by word recognition. In one aspect of the invention, a selectable transformation changes the given word to a differently spelled word having the same word root. In another, a selectable transformation changes a given word to one or more of its homonyms. In yet another, a selectable transformation changes the given word between a representation that spells the word with letters and one that does not. In one aspect of the invention a user can select to display a choice list of transformed words corresponding to a given recognized word and then select to have one of the listed transformed words replace the given word. In another aspect of the invention word recognition favors recognition of words corresponding to a user selected part of speech.

Claims

1. A computerized method of performing word recognition comprising:

receiving a word input signal representing one or more words;

performing word recognition upon the signal to produce one or more best scoring words corresponding to the word input signal;

treating one of said best scoring words as a currently selected word;

providing a user interface enabling a user to select from among a plurality of word transformation commands each associated with a different specific transformation for changing the currently selected word to a differently spelled word having the same word root;

responding to the user'"'"'s selection of one of the word transformation commands by causing the selected word transformation command'"'"'s associated transformation to be performed on the currently selected word.

2. The method as in claim 1 wherein at least one of the word transformation commands transforms the currently selected word to a different grammatical form.

3. The method as in claim 2 wherein at least one of the word transformation commands transforms the currently selected word to a different tense.

4. The method as in claim 2 wherein at least one of the word transformation commands transforms the currently selected word to a plural or singular form.

5. The method as in claim 2 wherein at least one of the word transformation commands transforms the currently selected word between a possessive and a non-possessive form.

6. The method as in claim 1 wherein at least one of the word transformation commands transforms the currently selected word by changing its ending to a given common word ending.

7. The method as in claim 1 wherein:

the word recognition produces a choice list of best scoring words, found by the recognizer to have the best probability of corresponding to the word signal; and

the user interface outputs the words of the choice list in user perceivable form; and

the user interface enables a user to select one of the words output on the choice list as to be treated said currently selected word and to have a selected one of the transformation commands performed upon said currently selected word, and to have the resulting transformed word produced as output of the recognition process.

8. The method as in claim 1 wherein

the word recognition is speech recognition performed on a telephone;

and

the user interface enables a user to select a selected one of the transformation commands by pressing a phone key.

9. A computerized method of performing word recognition comprising:

receiving a word input signal representing one or more words;

performing word recognition upon the signal to produce one or more best scoring words corresponding to the word input signal;

treating one of said best scoring words as a currently selected word;

providing a user interface enabling a user to select from among a plurality of word transformation commands, including a specific command to transform the currently selected word to a homonym of the currently selected word; and

responding to the user'"'"'s selection of the command to transform the currently selected word to a homonym by transforming a currently selected word to one or more of its homonyms.

10. A computerized method of performing word recognition comprising:

receiving a word input signal representing one or more words;

performing word recognition upon the signal to produce one or more best scoring words corresponding to the word input signal;

treating one of said best scoring words as a currently selected word;

providing a user interface enabling a user to select from among a plurality of word transformation commands, including a command for transforming the currently selected word between a letter and a non-letter form; and

responding to the user'"'"'s selection the command for transforming between a letter and a non-letter form by transforming a currently selected word between a spelled representation using one or more letters of the alphabet and a non-spelled representation that uses only characters that are not letters of the alphabet.

11. The method as in claim 10 wherein:

the word recognition produces a choice list of best scoring words, found by the recognizer to have a the best probability of corresponding to the signal; and

the user interface outputs the words of the choice list in user perceivable form; and

the user interface enables a user to select a one of words output on the choice list to be treated as said currently selected word and to have the transformation for changing between an alphabetic and a non-alphabetic representation performed upon that selected word, and to have the resulting transformed word representation produced as output of the recognition process.

12. The method as in claim 10 wherein said non-spelled form represents a word with one or more numerals.

13. The method as in claim 10 wherein said non-spelled form is a punctuation mark.

14. A computerized method of performing word recognition comprising:

receiving a word input signal representing one or more words;

performing word recognition upon the signal to produce one or more best scoring words corresponding to the word input signal;

treating one of said best scoring words as the currently selected word;

providing a user interface enabling a user to select a command to display a transformed word choice list;

responding to a user selection of said command by displaying a transformed word choice list produced by applying a plurality of transformations to the currently selected word and placing on the displayed transformed word choice list the transformed word, if any, that results from the application of each such transformation to the currently selected word;

providing a user interface enabling a user to select one of the transformed words in the displayed transformed word choice list; and

responding to the selection of such a transformed word by having the selected transformed word, instead of the currently selected word, be the output produced in correspondence with the recognition of the word input signal;

wherein each of said plurality of specific transformations is a specific transformation for changing a given currently selected word to a different word or form having a corresponding predefined relationship to the given word by transforming the given word;

to a word having a different grammatical form but the same word root;

to a homonym of the given word;

or

between a spelled representation using one or more letters of the alphabet and a non-spelled representation that uses only characters that are not letters of the alphabet.

15. A method a in claim 14 wherein:

the transformed word choice list is shown on a user scrollable display, with the choice list having more transformed words than fit on the display at one time so that only a sub-portion of the choice list is displayed at one time; and

the method responds to user input selecting to scroll the choice list up or down by moving the choice list relative to the display up or down, respectively, so as to change the portion of the choice list shown on the display.

16. A method as in claim 14 wherein the user interface:

places words output by the recognition process into a text; and

allows the user to select from among one or more words in the text the currently selected word for which the transformed word choice list is to be produced.

17. A method as in claim 14 wherein the user interface:

displays a choice list which includes a plurality of the best scoring word candidates from said word recognition; and

allows the user to select from among the words in the best scoring choice list the currently selected word for which a transformed word choice list is produced.

18. A method as in claim 14 wherein the words in the transformed word choice list include the one or more homonyms, if any, of the currently selected word for which the transformed word choice list is produced.

19. A method as in claim 14 wherein the words in the transformed word choice list include one or more different representations of the currently selected word for which the transformation choice list is produced.

20. A method as in claim 14 wherein the words in the transformed word choice list include one or more different grammatical forms, if any, of the currently selected word for which the transformation choice list is produced.