Method and apparatus for performing a name acquisition based on speech recognition

• US 6,208,965 B1
• Filed: 11/20/1997
• Issued: 03/27/2001
• Est. Priority Date: 11/20/1997
• Status: Expired due to Term

• Current Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Original Assignee: AT&T Corporation (AT&T, Inc.)
• Inventors: Rosinski, Richard R., Brown, Deborah W., Goldberg, Randy G.
• Primary Examiner(s): Hudspeth, David R.
• Assistant Examiner(s): MCFADDEN, SUSAN IRIS
• Application Number:
  US08/975,587
• Time in Patent Office: 1,223 Days
• Field of Search: 704/240, 704/241, 704/243, 704/244, 704/245, 704/246, 704/251, 704/256, 704/270, 704/272, 704/273, 704/275, 379/88, 455/563
• US Class Current: 704/246
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 17/24 the user being prompted to utter a password or a predefined phrase

Abstract

A method and apparatus for recognizing an input identifier on the basis of a set of comparison identifiers. After a user provides the input identifier according to a first form, the present invention provides a recognized identifier based on the input identifier. The present invention then generates a plurality of comparison identifiers on the basis of the recognized identifier. The user is then prompted to provide the input identifier again, but this time according to a second form that is different than the first form. A second recognized identifier is then generated on the basis of the input identifier provided according to the second form. If a match exists between the second recognized identifier and one of the comparison identifiers, the matched comparison identifier is selected as corresponding to the input identifier.

Claims

1. A method of recognizing an input identifier, comprising the steps of:

a) performing a first prompting operation to prompt a user at a first time in a recognition operation to provide the input identifier according to a first form;

b) providing a first recognized identifier on the basis of the input identifier;

the input identifier being provided by a user,

c) generating a plurality of comparison identifiers having a content derived from the first recognized identifier;

d) performing a second prompting operation to prompt the user at a second time in the recognition operation subsequent to the first time to provide the input identifier according to a second form, wherein the first form and the second form are different;

e) providing a second recognized identifier on the basis of the input identifier provided in step d);

f) determining which, if any, of the plurality of comparison identifiers matches the second recognized identifier; and

g) selecting the comparison identifier determined in step f) as corresponding to the input identifier.

2. The method according to claim 1, wherein the step a) comprises:

i) prompting the user to spell the input identifier.

3. The method according to claim 2, wherein the step d) comprises:

ii) prompting the user to pronounce the input identifier.

4. The method according to claim 3, wherein the input identifier comprises a name.

5. The method according to claim 1, wherein the step c) comprises:

i) determining, for each character position of the first recognized identifier, an alternative character set, each character of each alternative character set being associated with at least a predetermined probability of having been provided by the user in step a); and

ii) generating the plurality of comparison identifiers on the basis of at least each of the alternative character sets determined in step i).

6. The method according to claim 5, wherein each alternative character set determined in step i) is determined on the basis of a confusion matrix.

7. The method according to claim 5, further comprising:

iii) generating further comparison identifiers to be added to the plurality of comparison identifiers by performing at least one of a character addition operation, a character deletion operation, and a character substitution operation on at least a portion of the plurality of comparison identifiers generated in step ii); and

iv) eliminating from the plurality of comparison identifiers those comparison identifiers comprising non-words.

8. The method according to claim 1, wherein the step c) comprises:

i) providing a plurality of confusion sets, each confusion set including a different aggregation of characters;

ii) determining, for each character of the first recognized identifier, which confusion set includes which character of the first recognized identifier; and

iii) generating the plurality of comparison identifiers on the basis of at least each of the confusion sets determined in step ii).

9. The method according to claim 8, further comprising:

iv) generating further comparison identifiers to be added to the plurality of comparison identifiers by performing at least one of a character addition operation, a character deletion operation, and a character substitution operation on at least a portion of the plurality of comparison identifiers generated in step iii); and

v) eliminating from the plurality of comparison identifiers those comparison identifiers comprising non-words.

10. The method according to claim 1, wherein the content of the plurality of comparison identifiers is indeterminate prior to a performance of the recognition operation.

11. An apparatus for recognizing an input identifier, comprising:

a) first means for performing a first prompting operation to prompt a user at a first time in a recognition operation to provide the input identifier according to a first form;

b) first means for providing a first recognized identifier on the basis of the input identifier;

the input identifier being provided by a user,

c) first means for generating a plurality of comparison identifiers having a content derived from the first recognized identifier;

d) second means for performing a second prompting operation to prompt the user at a second time in the recognition operation subsequent to the first time to provide the input identifier according to a second form, wherein the second form is different than the first form;

e) second means for providing a second recognized identifier on the basis of the input identifier provided by the second means for prompting;

f) first means for determining which, if any, of the plurality of comparison identifiers matches the second recognized identifier; and

g) means for selecting the comparison identifier determined by the means for determining as corresponding to the input identifier.

12. The apparatus according to claim 11, wherein the first means for prompting comprises:

i) means for prompting the user to spell the input identifier.

13. The apparatus according to claim 12, wherein the second means for prompting comprises:

ii) means for prompting the user to pronounce the input identifier.

14. The apparatus according to claim 13, wherein the input identifier comprises a name.

15. The apparatus according to claim 11, wherein the first means for generating comprises:

i) second means for determining, for each character position of the first recognized identifier, an alternative character set, each character of each alternative character set being associated with at least a predetermined probability of having been provided by the user in step a); and

ii) second means for generating the plurality of comparison identifiers on the basis of at least each of the alternative character sets determined by the second means for determining.

16. The apparatus according to claim 15, wherein each alternative character set is determined on the basis of a confusion matrix.

17. The apparatus according to claim 15, further comprising:

iii) third generating means for generating further comparison identifiers to be added to the plurality of comparison identifiers by performing at least one of a character addition operation, a character deletion operation, and a character substitution operation on at least a portion of the plurality of comparison identifiers generated by the second generating means; and

iv) means for eliminating from the plurality of comparison identifiers those comparison identifiers comprising non-words.

18. The apparatus according to claim 11, wherein the first generating means comprises:

i) third means for providing a plurality of confusion sets, each confusion set including a different aggregation of characters;

ii) second means for determining, for each character of the first recognized identifier, which confusion set includes which character of the first recognized identifier; and

iii) second means for generating the plurality of comparison identifiers on the basis of at least each of the confusion sets determined by the second determining means.

19. The apparatus according to claim 18, further comprising:

iv) third means for generating further comparison identifiers to be added to the plurality of comparison identifiers by performing at least one of a character addition operation, a character deletion operation, and a character substitution operation on at least a portion of the plurality of comparison identifiers generated by the second means for generating; and

v) means for eliminating from the plurality of comparison identifiers those comparison identifiers comprising non-words.

20. An apparatus, comprising:

a processing device;

a recognizing device coupled to the processing device, the recognizing device including an input for receiving an input identifier;

the input identifier being provided by a user;

a database coupled to the processing device;

a grammar generation module coupled to the processing device; and

a confusion matrix memory coupled to the processing device.

21. The apparatus according to claim 20, further comprising:

a spelling rules table coupled to the processing device; and

a word change table coupled to the processing device.

22. An apparatus, comprising:

a processing device;

a recognizing device coupled to the processing device, the recognizing device including an input for receiving an input identifier;

a database coupled to the processing device;

a grammar generation module coupled to the processing device;

the input identifier being provided by a user;

a confusion matrix memory coupled to the processing device;

a confusion set generation module coupled to the processing device; and

a confusion set memory coupled to the processing device.

23. The apparatus according to claim 22, further comprising:

a spelling rules table coupled to the processing device; and

a word change table coupled to the processing device.