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

• US 8,280,013 B2
• Filed: 07/15/2008
• Issued: 10/02/2012
• Est. Priority Date: 12/15/2003
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: AT&T Intellectual Property I LP (AT&T, Inc.)
• Inventors: Bushey, Robert R., Martin, John M., Knott, Benjamin A.
• Primary Examiner(s): Anwah, Olisa
• Application Number:
  US12/173,553
• Publication Number:
  US 20080267365A1
• Time in Patent Office: 1,540 Days
• Field of Search: 379/88.01, 379/265.02, 379/265.01, 704/200
• US Class Current: 379/88.01
• CPC Class Codes: G10L 15/197 Probabilistic grammars, e.g. word n-grams
• CPC Class Codes: H04M 2201/40 using speech recognition
• 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. A method for identifying a routing destination in a service center, the method comprising:

prompting a caller to convey a request;

receiving the request as a natural language utterance from the caller;

comparing the natural language utterance to an action-object matrix, wherein the natural language utterance is searched for actions and objects in the action-object matrix;

when results of the comparing indicate that an action-object matrix pair was detected in the natural language utterance;

identifying a routing destination based upon the action-object matrix pair detected; and

routing the caller to the routing destination; and

when results of the comparing indicate that an action-object matrix pair was not detected and an action was detected in the natural language utterance;

querying the action-object matrix to identify objects related to the action detected; and

enabling the caller to select an identified related object.

2. The method of claim 1, further comprising:

when results of the comparing indicate that an action-object matrix pair was not detected, initiating a disambiguation dialog with the caller.

3. The method of claim 1, wherein the comparing further comprises:

using a statistical language modeling speech recognition utility to identify at least one of an action, an object and an action-object matrix pair.

4. A system for routing a caller to a service center based on a natural language request, the system comprising:

a processor having access to a memory media storing program instructions, executable by the processor, wherein the program instructions, when executed by the processor, perform a method comprising;

searching a natural language utterance indicative of a caller request for an action and an object;

responsive to identifying the action and the object;

identifying a task to be performed based on a task matrix, wherein the task matrix includes a plurality of action-object pairs including an action-object pair corresponding to the action identified and the object identified; and

directing the caller to a service center agent to perform the task;

responsive to identifying the object, but not identifying the action;

identifying, from the task matrix, a set of actions related to the object identified; and

directing the caller to indicate a selected action from the set of actions; and

responsive to identifying the action, but not identifying the object;

identifying from the task matrix a set of objects related to the action identified; and

directing the caller to indicate a selected object from the set of objects.

5. The system of claim 4, wherein the method includes:

responsive to the caller indicating the selected action, locating the service center agent based an action-object pair, in the task matrix, corresponding to the selected action and the object identified.

6. The system of claim 4, wherein the method includes providing, by the service center agent, automated assistance to the caller.

7. The system of claim 4, wherein the method includes providing, by the service center agent, technician assistance to the caller.

8. The system of claim 4, wherein directing the caller to the service center agent includes:

determining a routing destination for the service center agent using a look-up table containing service center agent routing destinations.

9. The system of claim 4, wherein the method includes:

indicating the task to the caller; and

prompting the caller to confirm the task indicated as the task identified.

10. The system of claim 4, wherein the searching of the natural language utterance includes:

identifying at least one of an action, an object, and an action-object pair with a statistical language modeling speech recognition utility.

11. A non-transitory computer-readable medium storing program instructions, executable by a processor, for routing callers to a service center destination, the program instructions, when executed by the processor, performing a method comprising:

searching a natural language utterance associated with a natural language caller request for an action and an object;

responsive to identifying the action and the object;

identifying, from a task matrix that includes action-object pairs, a task to be performed; and

directing the caller to a service center agent to perform the task;

responsive to identifying the object, but not identifying the action;

identifying, from the action-object matrix, a set of actions related to the object identified; and

directing the caller to indicate a selected action; and

responsive to identifying the action, but not identifying the object;

identifying, from the task matrix, a set of objects related to the action identified; and

directing the caller to indicate a selected object from the set of objects.

12. The computer readable medium of claim 11, wherein the method includes:

responsive to the caller indicating the selected action, locating the service center agent based on an action-object pair, in the task matrix, corresponding to the selected action and the object identified.

13. The computer readable medium of claim 11, wherein the method includes providing, by the service center agent, automated assistance to the caller.

14. The computer readable medium of claim 11, wherein the method includes providing, by the service center agent, technician assistance to the caller.

15. The computer readable medium of claim 11, wherein directing the caller to the service center agent includes:

determining a routing destination for the service center agent using a look-up table containing service center agent routing destinations.