Identifying a contact

• US 9,754,582 B2
• Filed: 02/14/2014
• Issued: 09/05/2017
• Est. Priority Date: 11/18/2013
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: Plumb, Graham, Swift, Mark
• Primary Examiner(s): Opsasnick, Michael N
• Assistant Examiner(s): BLANKENAGEL, BRYAN S
• Application Number:
  US14/180,957
• Publication Number:
  US 20150142442A1
• Time in Patent Office: 1,299 Days
• CPC Class Codes: G06F 40/20 Natural language analysis semantic analysis of natural language G06F40/30
• CPC Class Codes: G10L 15/08 Speech classification or search
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 17/22 Interactive procedures; Man-machine interfaces

Abstract

A method of identifying a contact in a communication system using voice input, the method comprising: receiving an input string of characters, the input string representing a contact and being normally unpronounceable by a human voice when spoken literally; performing at least one transforming step to transform at least one character of the input string to thereby generate a pronounceable name for the contact; and outputting the pronounceable name for use in establishing a communication event with the contact using voice input.

Claims

1. A method of identifying a contact in a communication system using voice input, the method comprising:

receiving an input string of characters, the input string representing a contact and being normally unpronounceable by a human voice when spoken literally;

displaying, on a user terminal, a message that the input string is going to be transformed into a pronounceable name for the contact;

performing at least one transforming step using the user terminal to transform at least one character of the input string to thereby generate the pronounceable name for the contact, the at least one transforming step being based at least in part on a region of the contact and performing regional sanitization based on the region to allow for cultural differences in the input string of characters by modifying the input string of characters to remove patterns of complexity and ambiguous string components;

returning a result including a number of character changes and a number of characters removed during the at least one transforming step; and

outputting the pronounceable name for use in establishing a communication event with the contact using voice input.

2. A method as claimed in claim 1, wherein the input string is displayed to a user on a display of the user terminal.

3. A method as claimed in claim 1, wherein the pronounceable name is stored in a grammar accessible by a speech recognition engine, whereby voice input of a pronounced version of the input string to the speech recognition engine locates a match of the pronounceable name.

4. A method as claimed in claim 2, wherein the pronounceable name is outputted to the display and displayed in association with the input string.

5. A method as claimed in claim 2, wherein the pronounceable name is outputted to the display in editable form as an autosuggested replacement for the input string.

6. A method as claimed in claim 1 further comprising outputting a quality parameter indicating the quality of the relationship between the input string and the pronounceable name.

7. A method as claimed in claim 6 further comprising discarding pronounceable names below a quality threshold, using the quality parameter.

8. A method as claimed in claim 1 wherein the at least one transforming step comprises multiple transforming steps, the output of one step being supplied to a next step.

11. A method as claimed in claim 1, wherein the input string representing the contact is written in “

leet speak”

.

12. A method as claimed in claim 1, wherein performing the at least one transforming step is dependent upon the contact'"'"'s locale and/or location.

13. A method as claimed in claim 6, wherein the at least one transforming step comprises multiple transforming steps, the output of one step being supplied to a next step and wherein the quality parameter is dependent upon a score for each step of the said multiple transforming steps.

9. A method as claimed in 8, wherein the at least one transforming step comprises at least one of:

replacing accents, stripping symbols, changing separators to whitespace, trimming leading/trailing numbers and symbols, replacing double vowel leet characters, replacing single vowel leet characters, stripping initials, normalising capitalisation, inserting whitespace in contiguous runs of characters between fragments distinguished by capitalisation, parsing suffixes and/or titles and/or nicknames and/or fullnames, and classifying said parsed suffixes and/or titles and parsing remaining characters of the input string into a first name and a fullname.

10. A method as claimed in claim 9, wherein the at least one transforming step comprises all of the transforming steps of claim 9.

14. A user device for identifying a contact in a communication system using voice input, the user device comprising a processor configured to execute code to:

receive an input string of characters, the input string representing a contact and being normally unpronounceable by a human voice when spoken literally;

display, on the user device, a message that the input string is going to be transformed into a pronounceable name for the contact;

perform at least one transforming step using the user device to transform at least one character of the input string to thereby generate the pronounceable name for the contact, the at least one transforming step being based at least in part on a region of the contact and performing regional sanitization based on the region to allow for cultural differences in the input string of characters by modifying the input string of characters to remove patterns of complexity and ambiguous string components;

display, on the user device, a message that the input string has been transformed into the pronounceable name; and

output the pronounceable name for use in establishing a communication event with the contact using voice input.

15. A user device as claimed in claim 14, wherein the pronounceable name is stored in a grammar accessible by a speech recognition engine, whereby voice input of a pronounced version of the input string to the speech recognition engine locates a match of the pronounceable name.

16. A user device as claimed in claim 14, wherein the input string is displayed to a user of the user device.

17. A user device as claimed in claim 16, wherein the pronounceable name is outputted to the display and displayed in association with the input string.

18. A user device as claimed in claim 16, wherein the pronounceable name is outputted to the display in editable form as an autosuggested replacement for the input string.

19. A user device as claimed in claim 14, which, when executed on the processor of the user device, is further configured to output a quality parameter indicating the quality of the relationship between the input string and the pronounceable name and to discard pronounceable names below a quality threshold, using the quality parameter.

20. A user terminal configured to identify a contact in a communication system using voice input, the user terminal comprising:

an input configured to receive an input string of characters, the input string representing a contact and being normally unpronounceable by a human voice when spoken literally;

a display displaying a message that the input string is going to be transformed into a pronounceable name for the contact, the message providing an option for not proceeding with transforming the input string into the pronounceable name;

transform logic configured to perform at least one transforming step to transform at least one character of the input string to thereby generate the pronounceable name for the contact, the at least one transforming step being based at least in part on a region of the contact and performing regional sanitization based on the region to allow for cultural differences in the input string of characters by modifying the input string of characters to remove patterns of complexity and ambiguous string components;

an output configured to output the pronounceable name for use in establishing a communication event with the contact using voice input and output a quality parameter indicating the quality of the relationship between the input string and the pronounceable name; and

a scoring module configured to discard pronounceable names below a quality threshold, using the quality parameter.

21. A user terminal as claimed in claim 20, wherein the input string of characters is received from a contact list stored on the user terminal.