System and method for speech recognition-enabled automatic call routing

• US 7,242,751 B2
• Filed: 12/06/2004
• Issued: 07/10/2007
• Est. Priority Date: 12/06/2004
• Status: Active Grant

• Current Assignee: Interactions, LLC
• Original Assignee: Sbc Knowledge Ventures, L.P.
• Inventors: Korth, Sarah, Martin, John Mills, Knott, Benjamin Anthony, Bushey, Robert R.
• Primary Examiner(s): Gauthier, Gerald
• Application Number:
  US11/005,494
• Publication Number:
  US 20060133587A1
• Time in Patent Office: 946 Days
• Field of Search: 379/88.03, 379/207.02, 379/88.01, 379/88.14, 704/221, 704/2, 704/252, 704/255, 704/270, 704/275, 701/202, 703/5
• US Class Current: 379/88.03
• CPC Class Codes: G10L 15/00 Speech recognition G10L17/00 takes precedence
• CPC Class Codes: H04M 2201/40 using speech recognition speech recognition per se G10L15/00
• CPC Class Codes: H04M 2203/2011 Service processing based on information specified by a party before or during a call, e.g. information, tone or routing selection
• CPC Class Codes: H04M 3/5166 in combination with interactive voice response systems or voice portals, e.g. as front-ends

Abstract

A system and method are disclosed for processing a call by receiving caller input in a speech format and utilizing phonemes to convert the speech input into word strings. The word strings are then converted into at least one object and at least one action. A synonym table is utilized to determine actions and objects. Objects generally represent nouns and adjective-noun combinations while actions generally represent verbs and adverb-verb combinations. The synonym table stores natural language phrases and their relationship with actions and objects. The actions and objects are utilized to determine a routing destination utilizing a routing table. The call is routed based on the routing table.

Claims

1. A method for processing a call comprising:

transforming speech input from a caller into a word string;

converting the word string into a detected object and a detected action;

determining, from a synonym table, a preferred object and a preferred action based on the detected object and the detected action;

determining a call destination based on the preferred object and the preferred action; and

routing the call to the call destination.

2. The method of claim 1, further comprising using phonemes to convert the speech input to the word string.

3. The method of claim 1, further comprising comparing text corresponding to the speech input to a list of word strings and assigning a probability to the word string.

4. The method of claim 1, wherein converting the word string into a detected object and a detected action comprises parsing the word string into the detected action and the detected object.

5. The method of claim 1, further comprising assigning a probability value to the word string wherein the probability value represents a probability that the word string matches an intent of the caller.

6. The method of claim 1, further comprising assigning a probability value to the preferred object wherein the probability value represents a probability that the preferred object represents an intent of the caller.

7. The method of claim 1, further comprising assigning a probability value to the preferred action wherein the probability value represents a probability that the preferred action represents an intent of the caller.

8. The method of claim 1, wherein the preferred action is one of a verb or an adverb-verb combination.

9. The method of claim 1, wherein the preferred object is one of a noun or an adjective-noun combination.

10. The method of claim 1, wherein the word string includes at least one informal or slang term.

11. The method of claim 1, further comprising identifying a geographic region of the caller, and wherein, converting the word string into the detected object and the detected action comprises converting the word string into the detected object and the detected action in dependence upon the identified geographic region of the caller.

12. A system for routing calls comprising:

a processor to convert speech input from a caller into a word string, and to convert the word sting into an object and an action;

a synonym table accessible by the processor to determine a preferred object based on the object and to determine a preferred action based ante action; and

an action-object routing module configured to determine a destination forte call based on the preferred object and the preferred action.

13. The system of claim 12, wherein the processor uses phonemes to convert the speech input to the word string.

14. The system of claim 12, wherein the processor assigns a probability to the word string.

15. The system of claim 12, wherein the processor parses the word string into the action and the object.

16. The system of claim 12, wherein the processor assigns a probability value to the word string and the probability value represents a probability that the word string represents an intent of the caller.

17. The system of claim 12, wherein the processor determines a probability value that represents a probability that the preferred object represents an intent of the caller.

18. The system of claim 12, wherein the processor determines a probability value that represents a probability that the preferred action represents an intent of the caller.

19. The system of claim 12, wherein the action is one of a verb or an adverb-verb combination.

20. The system of claim 12, wherein the object is one of a noun or an adjective-noun combination.

21. The system of claim 12, further comprising call routing system to route the call to the destination.

22. A method of call processing, the method comprising:

transforming speech input from a caller into one or more word strings;

converting each of the one or more word strings into an object and an action;

assigning a first probability value to the object and a second probability value to the action of each of the one or more word strings; and

determining a call destination from the object and the action based on the first and the second probability values.

23. The method of claim 22, wherein determining the call destination comprises:

identifying a dominant action and a dominant object of the one or more word strings based on the assigned first and second probability values; and

comparing the dominant action and the dominant object to predefined action-object pairs in a routing table to identify the call destination.

24. The method of claim 23, further comprising routing the call to the identified call destination.