Method and apparatus for locating a desired record in a plurality of records in an input recognizing telephone directory

• US 6,208,713 B1
• Filed: 12/05/1996
• Issued: 03/27/2001
• Est. Priority Date: 12/05/1996
• Status: Expired due to Term

• Current Assignee: Avaya Incorporated
• Original Assignee: Nortel Networks Limited (Nortel Networks Corporation)
• Inventors: Beaton, Brian Finlay, Rahrer, Timothy Joseph, Smith, Colin Donald
• Primary Examiner(s): Zele, Krista
• Assistant Examiner(s): Hoosain, Allan
• Application Number:
  US08/760,966
• Time in Patent Office: 1,573 Days
• Field of Search: 379/67, 379/91.01, 379/93.02, 379/93.03, 379/100.03, 379/88, 379/89, 379/133, 379/139, 379/67.1, 379/88.01, 379/88.02, 379/88.07, 379/88.08, 379/83, 379/85, 379/88.09, 379/111, 379/113
• US Class Current: 379/67.1
• CPC Class Codes: H04M 1/271 controlled by voice recognition
• CPC Class Codes: H04M 1/27457 Management thereof, e.g. manual editing of data
• CPC Class Codes: H04M 1/2746 Sorting, e.g. according to history or frequency of use
• CPC Class Codes: H04M 1/56 Arrangements for indicating or recording the called number at the calling subscriber's set
• CPC Class Codes: H04M 1/57 Arrangements for indicating or recording the number of the calling subscriber at the called subscriber's set at the operator set in a manual exchange H04M5/20
• CPC Class Codes: H04M 3/42 Systems providing special services or facilities to subscribers specially adapted for wireless communication networks H04W4/00
• CPC Class Codes: H04M 3/4931 Directory assistance systems

Abstract

A method and apparatus for locating a desired record amongst a plurality of records in a telephone directory involves associating a frequency field with each of the records, the frequency field being operable to store a frequency metric. A characteristic field is also associated with each of the records, the characteristic field being operable to store a characteristic representation of at least one identifying characteristic of a rendering associated with the record. An input representation of a rendering purporting to identify the desired record is produced and the input representation is compared with the contents of the characteristic fields of at least some of the records and confidence metrics are assigned to the records respectively, each of the confidence metrics representing a respective probability that a respective record is the desired record. A determination is then made as to which of the records have the first and second highest confidence metrics. The frequency fields of the records having the first and second highest confidence metrics are then read and the record with the highest frequency metric is identified as the desired record. Instruction codes stored on a computer-readable memory medium may be used to configure a general purpose computer to effect the method.

Claims

1. A method of locating a desired record amongst a plurality of records in a telephone directory, the method comprising:

a) associating a frequency field with each of said records, said frequency field being operable to store a frequency metric representing the frequency of use of said record, based on incoming and outgoing calls;

b) associating a characteristic field with each of said records, said characteristic field being operable to store a characteristic representation of at least one identifying characteristic of a rendering associated with said record;

c) producing an input representation of a rendering purporting to identify said desired record;

d) comparing said input representation with the contents of the characteristic fields of at least some of said records and assigning confidence metrics to said records respectively, each of said confidence metrics representing a respective probability that a respective record is the desired record;

e) determining which of said records have the first and second highest confidence metrics; and

f) selecting said desired record from said records having the first and second highest confidence metrics, as a function of said first and second highest confidence metrics and the frequency metrics of said records having the first and second highest confidence metrics.

2. A method as claimed in claim 1 further comprising increasing the contents of the frequency field of the desired record.

3. A method as claimed in claim 1 further comprising determining whether at least one of said first and second highest confidence metrics has a value greater than a first pre-defined value.

4. A method as claimed in claim 3 further comprising determining a difference between said first and second highest confidence metrics.

5. A method as claimed in claim 4 further comprising when said first highest confidence metric is greater than said first pre-defined value and said difference is greater than a second pre-defined value, selecting as said desired record the record associated with the first highest confidence metric.

6. A method as claimed in claim 4 further comprising including the step of when said first highest confidence metric is greater than said first pre-defined value, and said difference is less than a second pre-defined value, selecting as said desired record the record associated with the first highest frequency metric when said first highest frequency metric is greater than a third pre-defined value.

7. A method as claimed in claim 6 comprising, where said first highest frequency metric is less than said third pre-defined value, prompting a user to repeat the rendering.

8. A method as claimed in claim 7 wherein producing an input representation of a rendering comprises digitizing an utterance made by the user.

9. A method as claimed in claim 8 further comprising receiving an oral utterance from said user.

10. A method as claimed in claim 4 further comprising, when said first highest confidence metric is not greater than a fourth pre-defined value or said difference between said confidence metrics is not greater than a fifth pre-defined value, and when said difference in confidence metrics is less than said fifth pre-defined value, determining which of said records having the first and second highest confidence metrics has the first highest frequency metric and prompting a user to confirm that the record having the first highest frequency metric is the desired record.

11. A method as claimed in claim 10 further comprising, where the user confirms that the record having the first highest frequency metric is the desired record, selecting as the desired record, the record associated with the first highest frequency metric.

12. A method as claimed in claim 10 further comprising, where the user denies that the record having the first highest frequency metric is the desired record, prompting the user to confirm that the record having the second highest frequency metric is the desired record.

13. A method as claimed in claim 12 further comprising where the user confirms that the record having the second highest frequency metric is the desired record, selecting as the desired record, the record which has the second highest frequency metric.

14. A method as claimed in claim 12 further comprising where the user denies that the record having the second highest frequency metric is the desired record, prompting the user to repeat the rendering.

15. A method as claimed in claim 1 further comprising:

a) determining which of said records having the first and second highest confidence metrics has the highest value in its frequency field;

b) determining a figure of merit based on a weighted sum of the confidence metric associated with the record having the frequency field with the highest value, the difference between the first and second highest confidence metrics and the contents of the frequency field of said record having the highest value in its frequency field; and

c) determining whether said figure of merit is greater than a sixth pre-defined value and if so, selecting as said desired record said record having the highest value in its frequency field.

16. A method as claimed in claim 1 wherein said frequency field is further operable to store a frequency of success metric related to the number of times the related record is selected as the desired record.

17. An apparatus for locating a desired record amongst a plurality of records in a telephone directory, the apparatus comprising:

a) computer readable memory media operable to store said plurality of records, each of said records having a frequency field and a characteristic field, said frequency field being operable to store a frequency metric representing the frequency of use of said record, based on incoming and outgoing calls said characteristic field being operable to store a characteristic representation of at least one identifying characteristic of a rendering associated with said record;

b) a digitizer for producing an input representation of a rendering purporting to identify said desired record;

c) a processor; and

d) instruction codes operable to direct said processor to;

i) compare said input representation with the contents of the characteristic fields of at least some of said records and to assign confidence metrics to said records respectively, each of said confidence metrics representing a respective probability that a respective record is the desired record;

ii) determine which of said records have the first and second highest confidence metrics; and

iii) read the frequency fields of said records having the first and second highest confidence metrics and select said desired record from said records having the first and second highest confidence metrics, as a function of said first and second highest confidence metrics and the frequency metrics of said records having the first and second highest confidence metrics.

18. An apparatus as claimed in claim 17 further comprising instruction codes operable to direct said processor to increase the contents of the frequency field of the desired record.

19. An apparatus as claimed in claim 17 further comprising instruction codes operable to direct the processor to perform the step of determining whether at least one of said first and second highest confidence metrics have a value greater than a first pre-defined value.

20. An apparatus as claimed in claim 19 wherein said frequency field is further operable to store a frequency of success metric related to the number of times the related record is selected as the desired record.

21. An apparatus as claimed in claim 17 further comprising instruction codes operable to direct the processor to perform the step of where the user denies that the record having the second highest frequency metric is the desired record, prompting the user to repeat the rendering.

22. An apparatus as claimed in claim 21 further comprising instruction codes operable to direct the processor to perform the step of, where the user confirms that the record having the first highest frequency metric is the desired record, selecting as the desired record, the record associated with the first highest frequency metric.

23. An apparatus as claimed in claim 19 further comprising instruction codes operable to direct the processor to perform the step of when said first highest confidence metric is greater than said first pre-defined value, and said difference is less than a second pre-defined value, selecting as said desired record the record associated with the first highest frequency metric when said first highest frequency metric is greater than a third pre-defined value.

24. An apparatus as claimed in claim 23 further comprising instruction codes operable to direct the processor to perform the step of determining a difference between said first and second highest confidence metrics.

25. An apparatus as claimed in claim 24 further comprising instruction codes operable to direct the processor to perform the step of when said first highest confidence metric is greater than said first pre-defined value and said difference is greater than a second pre-defined value, select as said desired record the record associated with the first highest confidence metric.

26. An apparatus as claimed in claim 25 further comprising instruction codes operable to direct the processor to perform the step of when said first highest confidence metric is greater than said first pre-defined value, and said difference is less than a second predefined value, selecting as said desired record the record associated with the first highest frequency metric when said first highest frequency metric is greater than a third pre-defined value.

27. An apparatus as claimed in claim 21 further comprising instruction codes operable to direct the processor to perform the step of, where said first highest frequency metric is less than said third pre-defined value, prompting the user to repeat the rendering.

28. An apparatus as claimed in claim 27 further comprising instruction codes operable to direct the processor to perform the step of, when said first highest confidence metric is not greater than a fourth pre-defined value or said difference between said confidence metrics is not greater than a fifth pre-defined value, and when said difference in confidence metrics is less than said fifth pre-defined value, determining which of said records having the first and second highest confidence metrics has the first highest frequency metric and prompting the user to confirm that the record having the first highest frequency metric is the desired record.

29. An apparatus as claimed in claim 27 further comprising instruction codes operable to direct the processor to perform the steps of:

a) determining which of said records having the first and second highest confidence metrics has the highest value in its frequency field;

b) determining a figure of merit based on a weighted sum of the confidence metric associated with the record having the frequency field with the highest value, the difference between the first and second highest confidence metrics and the contents of the frequency field of said record having the highest value in its frequency field; and

c) determining whether said figure of merit is greater than a sixth pre-defined value and if so, selecting as said desired record said record having the highest value in its frequency field.

30. An apparatus as claimed in claim 29 further comprising instruction codes operable to direct the processor to perform the step of, where the user denies that the record having the first highest frequency metric is the desired record, prompting the user to confirm that the record having the second highest frequency metric is the desired record.

31. An apparatus as claimed in claim 29 further comprising instruction codes operable to direct the processor to perform the step of receiving an oral utterance from said user.

32. An apparatus as claimed in claim 17 wherein said frequency field is further operable to store a frequency of success metric related to the number of times the related record is selected as the desired record.

33. A computer-readable storage medium operable to store codes operable to direct a computer to locate a desired record amongst a plurality of records in a telephone directory, by:

a) associating a frequency field with each of said records, said frequency field being operable to store a frequency metric representing the frequency of use of said record, based on incoming and outgoing calls;

b) associating a characteristic field with each of said records, said characteristic field being operable to store a characteristic representation of at least one identifying characteristic of a rendering associated with said record;

c) producing an input representation of a rendering purporting to identify said desired record;

d) comparing said input representation with the contents of the characteristic fields of at least some of said records and assigning confidence metrics to said records respectively, each of said confidence metrics representing a respective probability that a respective record is the desired record;

e) determining which of said records have the first and second highest confidence metrics; and

f) selecting said desired record from said records having the first and second highest confidence metrics, as a function of said first and second highest confidence metrics and the frequency metrics of said records having the first and second highest confidence metrics.

34. A computer-readable storage medium as claimed in claim 33 further comprising codes operable to direct the computer to perform the step of increasing the contents of the frequency field of the desired record.

35. A computer-readable storage medium as claimed in claim 33 further comprising codes operable to direct the computer to perform the step of determining whether at least one of said first and second highest confidence metrics has a value greater than a first pre-defined value.

36. A computer-readable storage medium as claimed in claim 33 further comprising codes operable to direct the computer to perform the step of determining a difference between said first and second highest confidence metrics.

37. A computer-readable storage medium as claimed in claim 33 further comprising codes operable to direct the computer to perform the step of when said first highest confidence metric is greater than said first pre-defined value and said difference is greater than a second pre-defined value, selecting as said desired record that record associated with the first highest confidence metric.

38. A computer-readable storage medium as claimed in claim 33 further comprising codes operable to direct the computer to perform the step of when said first highest confidence metric is greater than said first pre-defined value, and said difference is less than a second pre-defined value, selecting as said desired record the record associated with the first highest frequency metric when said first highest frequency metric is greater than a third pre-defined value.

39. A computer-readable storage medium as claimed in claim 38 further comprising codes operable to direct the computer to perform the step of, where said first highest frequency metric is less than said third pre-defined value, prompting the user to repeat the rendering.

40. A computer-readable storage medium as claimed in claim 39 further comprising codes operable to direct the computer to perform the step of digitizing an utterance made by the user.

41. A computer-readable storage medium as claimed in claim 40 further comprising codes operable to direct the computer to perform the step of receiving an oral utterance from said user.

42. A computer-readable storage medium as claimed in claim 38 further comprising codes operable to direct the computer to perform the step of, when said first highest confidence metric is not greater than a fourth pre-defined value or said difference between said confidence metrics is not greater than a fifth pre-defined value, and when said difference in confidence metrics is less than said fifth pre-defined value, determining which of said records having the first and second highest confidence metrics is associated with the first highest frequency metric and prompting the user to confirm that the record having the first highest frequency metric is the desired record.

43. A computer-readable storage medium as claimed in claim 42 further comprising codes operable to direct the computer to perform the step of, where the user confirms that the record having the first highest frequency metric is the desired record, selecting as the desired record, the record associated with the first highest frequency metric.

44. A computer-readable storage medium as claimed in claim 43 further comprising codes operable to direct the computer to perform the step of, where the user denies that the record having the first highest frequency metric is the desired record, prompting the user to confirm that the record having the second highest frequency metric is the desired record.

45. A computer-readable storage medium as claimed in claim 44 further comprising codes operable to direct the computer to perform the step of where the user confirms that the record having the second highest frequency metric is the desired record, selecting as the desired record, the record which has the second highest frequency metric and incrementing the contents of the frequency field of said desired record.

46. A computer-readable storage medium as claimed in claim 44 further comprising codes operable to direct the computer to perform the step of where the user denies that the record having the second highest frequency metric is the desired record, prompting the user to repeat the rendering.

47. A computer-readable storage medium as claimed in claim 33 further comprising codes operable to direct the computer to perform the steps of:

a) determining which of said records having the first and second highest confidence metrics has the highest value in its frequency field;

b) determining a figure of merit based on a weighted sum of the confidence metric associated with the record having the frequency field with the highest value, the difference between the first and second highest confidence metrics and the contents of the frequency field of said record having the highest value in its frequency field; and

c) determining whether said figure of merit is greater than a sixth pre-defined value and if so, selecting as said desired record said record having the highest value in its frequency field.

48. A computer-readable storage medium as claimed in claim 33 wherein said frequency field is further operable to store a frequency of success metric related to the number of times the related record is selected as the desired record.

49. A method of locating a desired record amongst a plurality of records in a telephone directory, the method comprising:

a) associating a frequency metric with each of said records, said frequency metric representing the incoming and outgoing calls;

b) associating a characteristic representation with each of said records, said characteristic representation identifying a rendering;

c) receiving an input representation of a rendering to identify said desired record;

d) comparing said input representation with the characteristic representation for at least some of said records and assigning confidence metrics to said at least some of said records, each of said confidence metrics representing a probability that a respective record is the desired record; and

e) selecting said desired record as a function of said confidence metrics and the frequency metrics.