System and method for speech-enabled call routing
First Claim
Patent Images
1. A method of processing a call comprising:
- receiving speech input from a caller;
using phonemes to convert the received speech input into one or more word strings in accordance with an acoustic model;
converting the one or more word strings into one or more actions and one or more objects in accordance with a semantic model;
assigning an action confidence level to each action of the one or more actions, wherein the action confidence level represents a probability that a particular action reflects an intended caller action;
assigning an object confidence level to each object of the one or more objects, wherein the object confidence level represents a probability that a particular object reflects an intended caller object;
determining an action-object pair that represents a highest confidence level action and a highest confidence level object;
determining a desired call destination based on the action-object pair using a routing table;
determining whether the caller is in compliance with a first business rule related to a maximum amount of time for a voice message box of the caller to be full;
routing the call to a second call destination when the caller is not in compliance with the first business rule, wherein the caller is prompted to comply with the first business rule at the second call destination; and
upon compliance with the first business rule, routing the call to the desired call destination.
6 Assignments
0 Petitions
Accused Products
Abstract
A method of processing a call is disclosed. The method can transform speech input from a caller of a call into text and convert the text into an object and an action. The method determines a call destination based on the object and the action. The method can route the call to a destination when a caller is not in compliance with at least one business rule. The method can further route the call to the call destination when the caller is in compliance.
-
Citations
12 Claims
-
1. A method of processing a call comprising:
-
receiving speech input from a caller; using phonemes to convert the received speech input into one or more word strings in accordance with an acoustic model; converting the one or more word strings into one or more actions and one or more objects in accordance with a semantic model; assigning an action confidence level to each action of the one or more actions, wherein the action confidence level represents a probability that a particular action reflects an intended caller action; assigning an object confidence level to each object of the one or more objects, wherein the object confidence level represents a probability that a particular object reflects an intended caller object; determining an action-object pair that represents a highest confidence level action and a highest confidence level object; determining a desired call destination based on the action-object pair using a routing table; determining whether the caller is in compliance with a first business rule related to a maximum amount of time for a voice message box of the caller to be full; routing the call to a second call destination when the caller is not in compliance with the first business rule, wherein the caller is prompted to comply with the first business rule at the second call destination; and upon compliance with the first business rule, routing the call to the desired call destination. - View Dependent Claims (2, 3, 4, 5, 6)
-
-
7. A speech-enabled call routing system comprising:
-
a processor; a network interface to receive speech input from a caller; an acoustic model executable by the processor to use phonemes to convert the received speech input into one or more word strings; a semantic model executable by the processor to convert the one or more word strings into one or more actions and one or more objects; an action object routing module executable by the processor to; assign an action confidence level to each action of the one or more actions, wherein the action confidence level represents a probability that a particular action reflects an intended caller action; and assign an object confidence level to each object of the one or more objects, wherein the object confidence level represents a probability that a particular object reflects an intended caller object; determine an action-object pair that represents a highest confidence level action and a highest confidence level object; determine a desired call destination based on the action-object pair using a routing table; a business rule module executable by the processor to; determine whether the caller is in compliance with a business rule related to a maximum amount of time for a voice message box of the caller to be full; wherein the speech-enabled routing system routes the call to a second destination when the caller is not in compliance with the business rule, wherein the caller is prompted to comply with the business rule at the second call destination; and wherein the speech-enabled routing system routes the call to the desired call destination upon compliance with the business rule. - View Dependent Claims (8, 9, 10, 11, 12)
-
Specification