Granted Patent B2

US 8,112,282 · App. 12/634,434 · Granted Feb 7, 2012

Abstract

Claims (50)

1. A method for generating prompts for a speech-enabled application, the speech-enabled application operable to receive communications from users and to communicate prompts to elicit responses from the users indicating a purpose for a communication, the method comprising:

determining, by a computer, a plurality of prompt alternatives to evaluate, each prompt alternative including an alternative prompt and an associated evaluation period;

presenting the alternative prompt to users during the evaluation period;

recording user responses to the alternative prompt;

analyzing, based on a performance criteria, user responses to the alternative prompt; and

implementing one of the plurality of prompt alternatives based on the analyzing of the user responses.

2. The method of claim 1 , wherein:

the plurality of prompt alternatives include a combination prompt comprising multiple prompts; and

the method further comprises determining an order in which to present the multiple prompts.

3. The method of claim 1 , wherein the evaluation period comprises a period during which a selected number of communications are received.

4. The method of claim 1 , wherein, when the user responses are recorded, direct routes resulting from the responses to each of the prompt alternatives are recorded; and

wherein analyzing the user responses includes determining which alternative prompt has the highest percentage of direct routes.

5. The method of claim 1 , wherein the computer apparatus is operable to match the user responses with an action-object based on salient terms, in the user responses, associated with the action-object;

wherein, when the user responses are recorded, information indicating a number of user responses that match with an action-object is recorded; and

wherein analyzing the user responses includes determining a prompt alternative having the highest percentage of responses that match with an action-object.

6. The method of claim 1 , wherein the computer apparatus is implemented in a call center.

7. A computer-implemented system for optimizing prompts for a speech-enabled application, the speech-enabled application operable to receive communications from a plurality of users and to communicate prompts to users to elicit user responses indicating a purpose of the communication, the system comprising:

a processor configured to access a computer readable media, wherein the computer readable media includes program instructions, executable by the processor, the program instructions including instructions to:

provide a test control module configured to:

determine a plurality of prompt alternatives to evaluate, each prompt alternative comprising at least one prompt; and

determine evaluation periods for each of the prompt alternatives;

provide a dialog module configured to present each of the prompt alternatives to a user during a corresponding evaluation period; and

provide a test analysis module configured to:

record user responses received for each of the prompt alternatives;

analyze the results for each of the prompt alternatives based on a performance criteria; and

implement one of the prompt alternatives based on the analysis of the results.

8. The system of claim 7 , wherein:

a prompt alternative includes a combination of prompts; and

the test control is further configured to determine an order in which to present the combination of prompts.

9. The system of claim 7 , wherein the evaluation period comprises a period during which a selected number of communications are received.

10. The system of claim 7 , wherein, when the test analysis module records the user responses, information indicative of a number of direct routes resulting from the user responses to each of the prompt alternatives is recorded.

11. The system of claim 10 , wherein, when the test analysis module analyzes the results for each of the prompt alternatives, a prompt alternative having a highest percentage of direct routes is determined.

12. The system of claim 7 , wherein the program instructions include instructions to:

match the user responses with an action-object based on salient terms in the user responses that are associated with the action-object; and

wherein when the test analysis module records the results of the responses, a number of responses that match with an action-object is determined.

13. The system of claim 12 , wherein analyzing the recorded results for each prompt alternative comprises determining which prompt alternative has the highest percentage of responses that the speech-enabled application is able to match with an action-object.

14. A non-transitory computer readable media including program instructions for optimizing prompts for a speech-enabled application, the speech-enabled application operable to receive a communication from each of a plurality of users and to communicate prompts to each user to elicit responses from each user indicating a purpose of the communication, the program instructions including instructions to:

determine a plurality of prompt alternatives to evaluate, each prompt alternative comprising at least one prompt and being respectively associated with an evaluation period;

present each prompt alternative to a user during the evaluation period;

analyze responses from the user based on a performance criteria, wherein the responses are respectively recorded after each prompt alternative is presented to the user; and

implement one of the prompt alternatives based on the instructions executable to analyze the responses.

15. The computer readable media of claim 14 , wherein the prompt alternatives comprise a combination of prompts, and including instructions executable to:

determine an order in which to present a prompt in the combination.

16. The computer readable media of claim 14 , wherein the evaluation period comprises a period during which a selected number of communications are received.

17. The computer readable media of claim 14 , wherein when the responses are recorded, a number of direct routes that result from the responses to each prompt alternative are recorded; and

wherein the instructions to analyze include instructions to determine which prompt alternative has the highest percentage of direct routes.

18. The computer readable media of claim 14 , including instructions executable to match the responses with an action-object based on salient terms; and

wherein when the responses are recorded, information indicative of a number of responses that correspond with the action-object is recorded.

19. The computer readable media of claim 18 , wherein the instructions to analyze the responses for each prompt alternative include instructions to determine a prompt alternative having the highest percentage of responses that correspond with an action-object.

20. The computer readable media of claim 14 , wherein the computer readable media is implemented in a call center computing apparatus.

Assignments (18)