METHODS AND APPARATUS TO PERFORM SPEECH REFERENCE ENROLLMENT

• US 20080015858A1
• Filed: 07/09/2007
• Published: 01/17/2008
• Est. Priority Date: 05/27/1997
• Status: Abandoned Application

• Current Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Original Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Inventors: Bossemeyer, Robert
• Application Number:
  US11/774,987
• Publication Number:
  US 20080015858A1
• Time in Patent Office: Days
• US Class Current: 704/243
• CPC Class Codes: G10L 15/07 to the speaker
• CPC Class Codes: G10L 17/04 Training, enrolment or model building
• CPC Class Codes: G10L 2015/0631 Creating reference templates; Clustering
• CPC Class Codes: G10L 2015/0636 Threshold criteria for the updating
• CPC Class Codes: G10L 2015/0638 Interactive procedures
• CPC Class Codes: H04M 2201/40 using speech recognition
• CPC Class Codes: H04M 3/382 using authorisation codes or passwords
• CPC Class Codes: H04M 3/493 Interactive information services, e.g. directory enquiries ; Arrangements therefor, e.g. interactive voice response [IVR] systems or voice portals

Abstract

A speech reference enrollment method involves the following steps: (a) requesting a user speak a vocabulary word; (b) detecting a first utterance (354); (c) requesting the user speak the vocabulary word; (d) detecting a second utterance (358); (e) determining a first similarity between the first utterance and the second utterance (362); (f) when the first similarity is less than a predetermined similarity, requesting the user speak the vocabulary word; (g) detecting a third utterance (366); (h) determining a second similarity between the first utterance and the third utterance (370); and (i) when the second similarity is greater than or equal to the predetermined similarity, creating a reference (364).

Claims

1-22. -22. (canceled)

23. A method, comprising:

receiving a first utterance of a word;

receiving a second utterance of the word;

when a number of voiced speech frames associated with the second utterance is greater than a threshold, determining a first similarity between the first utterance and the second utterance; and

when the first similarity is greater than or equal to a similarity threshold, storing a reference for the word.

24. A method as defined in claim 23, wherein determining the first similarity between the first utterance and the second utterance comprises determining a similarity between a first plurality of features associated with the first utterance and a second plurality of features associated with the second utterance.

25. A method as defined in claim 23, further comprising:

when the first similarity is less than the similarity threshold, requesting a user to speak a third utterance of the word;

determining a second similarity between the first utterance and the third utterance; and

when the second similarity is greater than or equal to the similarity threshold, storing the reference for the word.

26. A method as defined in claim 23, further comprising determining the number of voiced speech frames by estimating the number of voiced speech frames.

27. A method as defined in claim 23, further comprising requesting the user repeat the word when the number of voiced speech frames is less than the threshold.

28. A method as defined in claim 23, further comprising:

determining a signal to noise ratio of the first utterance; and

when the signal to noise ratio is less than a threshold signal to noise ratio, increasing a gain of a voice amplifier.

29. A method as defined in claim 23, further comprising determining an amplitude histogram of the first utterance.

30. A method as defined in claim 23, further comprising retrieving the reference and verifying a speaker based on the reference.

31. A method, comprising:

receiving a first utterance of a word;

receiving a second utterance of the word;

when the duration of the second utterance is greater than or equal to a first duration or less than or equal to a second duration, storing a reference for the word.

32. A method as defined in claim 31, further comprising:

when the duration of the second utterance is greater than or equal to the first duration or less than or equal to the second duration, determining a first similarity between the first utterance and the second utterance; and

when the first similarity is greater than or equal to a similarity threshold, storing the reference.

33. A method as defined in claim 32, further comprising:

when the first similarity is less than the similarity threshold, requesting the user speak the word;

determining a second similarity between the first utterance and a third utterance; and

when the second similarity is greater than or equal to the similarity threshold, storing the reference.

34. A method as defined in claim 33, further comprising:

determining a third similarity between the second utterance and the third utterance; and

when the third similarity is greater than or equal to the similarity threshold, storing the reference.

35. A method as defined in claim 31, further comprising:

determining if the first utterance exceeds an amplitude threshold within a time period; and

when the first utterance does not exceed the amplitude threshold within the time period, requesting the user re-speak the word.

36. A method as defined in claim 31, further comprising:

associating a start time with a point at which a first amplitude of the first utterance is greater than an amplitude threshold;

associating an end time with a point at which a second amplitude of the first utterance is less than the amplitude threshold; and

determining the duration as a difference between the end time and the start time.

37. A method as defined in claim 35, further comprising:

determining a signal to noise ratio of the first utterance; and

when the signal to noise ratio is less than a threshold signal to noise ratio, increasing a gain of a voice amplifier.

38. A method as defined in claim 31, further comprising retrieving the reference and verifying a speaker based on the reference.

39. A machine accessible storage medium having instructions stored thereon that, when executed, cause a machine to:

receive a first utterance of a word;

receive a second utterance of the word;

when the duration of the second utterance is greater than or equal to a first duration or less than or equal to a second duration, store a reference for the word.

40. A machine accessible storage medium as defined in claim 39 having instructions stored thereon that, when executed, cause the machine to:

when the duration of the second utterance is greater than or equal to the first duration or less than or equal to the second duration;

determine a first similarity between the first utterance and the second utterance; and

when the first similarity is greater than or equal to a similarity threshold, store the reference.

41. A machine accessible storage medium as defined in claim 40 having instructions stored thereon that, when executed, cause the machine to:

when the first similarity is less than the similarity threshold;

request the user speak the word;

determine a second similarity between the first utterance and a third utterance; and

when the second similarity is greater than or equal to the similarity threshold, store the reference.

42. A machine accessible storage medium as defined in claim 41 having instructions stored thereon that, when executed, cause the machine to:

determine a third similarity between the second utterance and the third utterance; and

when the third similarity is greater than or equal to the predetermined similarity, store the reference.

43. A machine accessible storage medium as defined in claim 39 having instructions stored thereon that, when executed, cause the machine to:

determine if the first utterance exceeds an amplitude threshold within a time period; and

when the first utterance does not exceed the amplitude threshold within the time period, request the user re-speak the word.

44. A machine accessible storage medium as defined in claim 39 having instructions stored thereon that, when executed, cause the machine to:

associate a start time with a point at which a first amplitude of the first utterance is greater than an amplitude threshold;

associate an end time with a point at which a second amplitude of the first utterance is less than the amplitude threshold; and

determine the duration as a difference between the end time and the start time.

45. A machine accessible storage medium as defined in claim 39 having instructions stored thereon that, when executed, cause the machine to:

determine a signal to noise ratio of the first utterance; and

when the signal to noise ratio is less than a threshold signal to noise ratio, increase a gain of a voice amplifier.

46. A machine accessible storage medium as defined in claim 39 having instructions stored thereon that, when executed, cause the machine to determine an amplitude histogram of the first utterance.

47. A machine accessible storage medium as defined in claim 39 having instructions stored thereon that, when executed, cause the machine to retrieve the reference and verify a speaker based on the reference.