System, method and software for a speech-enabled call routing application using an action-object matrix

• US 20050132262A1
• Filed: 12/15/2003
• Published: 06/16/2005
• Est. Priority Date: 12/15/2003
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Sbc Knowledge Ventures, L.P.
• Inventors: Bushey, Robert R., Knott, Benjamin A., Martin, John M.
• Granted Patent: US 7,415,101 B2
• Time in Patent Office: Days
• US Class Current: 714/760
• CPC Class Codes: G10L 15/197 Probabilistic grammars, e.g. word n-grams
• CPC Class Codes: H04M 2201/40 using speech recognition speech recognition per se G10L15/00
• CPC Class Codes: H04M 3/42059 Making use of the calling party identifier
• CPC Class Codes: H04M 3/42204 Arrangements at the exchange for service or number selection by voice at the terminal H04M1/27
• CPC Class Codes: H04M 3/523 with call distribution or queueing
• CPC Class Codes: H04M 3/527 Centralised call answering arrangements not requiring operator intervention
• CPC Class Codes: H04M 7/1215 where a cable TV network is used as an access to the PSTN/ISDN
• CPC Class Codes: H04M 7/1255 where the switching fabric and the switching logic are decomposed such as in Media Gateway Control

Abstract

A system, method and software for facilitating a speech-enabled call routing application using an action-object matrix is disclosed. In operation, a natural language user utterance may be evaluated to identify an action and object available in an action-object matrix indicating transactions or operations available to a user. Depending upon the contents of the natural language user utterance, additional prompts and/or a disambiguation dialogue may be effected to elicit an available action-object combination selection from the user. Following identification of an action-object combination from the natural language user utterance, the action-object matrix may cooperate with a look-up table to identify an appropriate use routing destination. Following identification of an appropriate routing destination, the user connection may be routed to a service agent or module configured to facilitate the user selected transaction as indicated by the action-object combination.

Claims

1. Software for routing calls in a call center based on a transaction request obtained from a natural language caller utterance, the software embodied in computer readable media and when executed operable to:

evaluate the natural language caller utterance in accordance with a statistical language modeling speech recognition utility;

determine whether the natural language caller utterance includes an action;

determine whether the natural language caller utterance includes an object;

if the natural language caller utterance includes only an action, elicit a natural language caller utterance identifying an object;

if the natural language caller utterance includes only an object, elicit a natural language caller utterance identifying an action;

if neither an action nor an object are included in the natural language caller utterance, prompt the caller for a natural language utterance identifying an action and an object;

once an action and an object have been identified, locate an intersection of the action and the object in an action-object matrix;

determine a routing destination from a look-up table associated with the action-object matrix intersection; and

direct the caller to the routing destination.

2. A method for identifying a routing destination in a service center, comprising:

prompting a user to convey a request;

receiving a natural language utterance from the user;

comparing the natural language utterance to an action-object matrix;

identifying a routing destination based upon results of the natural language utterance to action-object matrix comparison; and

routing the user to the routing destination.

3. The method of claim 2, further comprising:

querying the action-object matrix to identify related objects if only an action and not an object is included in the natural language utterance; and

prompting the user for selection of a related object.

4. The method of claim 2, further comprising:

querying the action-object matrix to identify related actions if an object and not an action is included in the natural language utterance; and

prompting the user for selection of a related action.

5. The method of claim 2, further comprising initiating a disambiguation dialog with the user where an action-object combination cannot be found in the action-object matrix.

6. The method of claim 2, further comprising identifying at least one of an action, an object and an action-object combination using a statistical language modeling speech recognition utility.

7. A system for routing a service center user based on a natural language request, comprising:

at least one processor;

memory operably associated with the at least one processor;

a program of instructions storable in the memory and executable by the processor, the program of instructions operable to identify a task to be performed from a natural language user request and a task matrix and to direct the user to a service center agent for performance of the task.

8. The system of claim 7, further comprising the program of instructions operable to:

identify an action-object combination from the natural language user request; and

locate a service agent to perform the task based on the identified action-object combination and the task matrix.

9. The system of claim 7, further comprising the service center agent operable to provide automated user assistance in performance of the task.

10. The system of claim 7, further comprising the service center agent operable to provide technician assisted performance of the task.

11. The system of claim 7, further comprising the task matrix including a plurality of available service center actions each cross-referenced with one or more objects creating action-object combinations and where the action-object combinations define tasks available from a service center agent.

12. The system of claim 11, further comprising the program of instructions operable to reference a look-up table containing service center agent routing destinations in response to a match between an action and an object in the natural language utterance and an action-object combination in the action-object matrix.

13. The system of claim 7, further comprising the program of instructions operable to identify at least one of an action or an object in the natural language utterance to identify the task to be performed.

14. The system of claim 7, further comprising the program of instructions operable to prompt the user for at least one additional natural language user request in response to identifying an action and no object in the natural language user utterance.

15. The system of claim 7, further comprising the program of instructions operable to prompt the user for at least one additional natural language user request in response to identifying an object and no action in the natural language user utterance.

16. The system of claim 7, further comprising the program of instructions operable to prompt the user for confirmation of the identified task requested.

17. Software for routing users to an appropriate service center destination, the software stored in computer readable media and when executed operable to:

match a transaction request derived from a natural language utterance to a transaction option in a transaction option matrix; and

facilitate connection between the user and a service module operable to effect processing of the requested transaction.

18. The software of claim 17, further operable to initiate a disambiguation dialog with a user in response to a failure to match a transaction request derived from the natural language utterance to a transaction in the transaction option matrix.

19. The software of claim 17, further operable to derive an action-object combination from the natural language utterance.

20. The software of claim 19, further operable to:

locate an action-object intersection in the transaction option matrix matching the derived action-object combination; and

reference a look-up table associated with the transaction option matrix, the look-up table containing destination information for a service module operable to effect the transaction option associated with the action-object intersection.

21. The software of claim 19, further operable to:

query an action-object matrix in response to derivation of an action without an object to identify available objects associated with the derived action; and

prompt the user for a natural language utterance selection of an available object associated with the derived action.

22. The software of claim 19, further operable to:

query an action-object matrix in response to derivation of an object without an action to identify available action associated with the derived object; and

prompt the user for a natural language utterance selection of an available action associated with the derived object.