Voice response system

• US RE45,096 E1
• Filed: 07/16/2009
• Issued: 08/26/2014
• Est. Priority Date: 04/19/2001
• Status: Active Grant

• Current Assignee: British Telecommunications PLC (BT Group PLC)
• Original Assignee: British Telecommunications PLC (BT Group PLC)
• Inventors: Harrison, Michael Andrew, Popay, Paul Ian, Watton, Neil Lewis
• Primary Examiner(s): Phan, Joseph T
• Application Number:
  US12/458,605
• Time in Patent Office: 1,867 Days
• Field of Search: 379/88.23
• US Class Current: 379/88.16
• CPC Class Codes: N/A

Abstract

With interactive voice response services, it can be frustrating for a user to become stuck in a dialogue where the same question is asked repetitively. Here the wording of questions used by the system are varied throughout the dialogue depending upon how many times a user has visited a particular dialogue state during the call history. Furthermore the wording of the question is also varied in dependence upon the way in which the question was asked the last time the user was in a particular dialogue state.

Claims

1. A method of operating an interactive voice response system to engage in a current dialogue with a user, said system having a dialogue model comprising

a plurality of states and a plurality of interconnecting edges;

a current state;

user dialogue data indicating for a user a total number of visits to a state;

in which the wording of a message for sending to the user at a state repeated in said current dialogue, is selected in dependence upon

a) the current state; and

b) the number of times the current state has been repeated during the current dialogue; and

c) the total number of times the current state has been repeated during one or more previous dialogues with said user; and

repeating a) through c) for multiple ones of said plurality of states.

2. A method according to claim 1 in which the wording of said message is selected in dependence on further data indicating whether the user has previously been at the current stage during the current dialogue and upon data indicating the wording of the message which was selected for the most recent visit to the current state.

6. A method according to claim 1 in which the message wording is selected in dependent on further data indicating whether the user has been at the current stage during the current dialogue and upon data indicating the message wording which was selected the previous time the user was at the current stage.

16. A method as in claim 1, wherein the user enters a unique identifier and a personal identification number which must match an entry in a user access database before the user can access the system.

17. A method as in claim 2, wherein the user enters a unique identifier and a personal identification number which must match an entry in a user access database before the user can access the system.

3. An interactive voice response system having a dialogue model comprising:

a plurality of states and a plurality of interconnecting edges;

a current state;

user dialogue data indicating for a user a total number of visits to a state;

the system further comprising message wording selection means, for selecting the wording of a message for sending to the user at a state repeated in said current dialogue, in dependence upon

a) the current state; and

b) the number of times the current state has been repeated during the current dialogue; and

c) the total number of times the current state has been repeated during one or more previous dialogues; and

repeating a) through c) for multiple ones of said plurality of states.

4. A system according to claim 3, wherein the message wording selection means is further operable to select the message wording in dependence on further data indicating whether the user has visited the current state during the current dialogue and upon data indicating the message wording which was selected for the most recent visit to the current state.

8. A system according to claim 3, wherein the message wording selection means is further operable to select the message wording in dependence on further data indicating whether the user has earlier visited the current stage during the current dialogue and upon data indicating the message wording which was selected on the most recent occasion on which the user was at the current stage.

10. A system as in claim 3, wherein the user enters a unique identifier and a personal identification number which must match an entry in a user access database before the user can access the system.

11. A system as in claim 10, wherein the user enters the unique identifier and personal identification number using a telephone keypad.

12. A system as in claim 10, wherein the user enters the unique identifier and personal identification number by speaking into a telephone.

5. A method of operating an interactive voice response system to engage in a current dialogue with a user, said interactive voice response system storing a dialogue model comprising

a plurality of states and a plurality of interconnecting edges,

a current state,

user dialogue data indicating for a user a total number of visits to a state in the current dialogue and a total number of visits in one or more previous dialogues;

wherein each state represents a stage of the dialogue and each interconnecting edge represents a transition between one stage of the dialogue and a subsequent stage,

said method comprising the steps of

finding the number of times the current stage has been repeated in the current and previous dialogues from said user dialogue data; and

selecting the wording to be output by said interactive voice response system at a repeated stage of the current dialogue in dependence upon

a) the current stage in the dialogue; and

b) the number of times the current stage has been repeated, and

c) the total number of times the user has been at this stage during one or more previous and

repeating a) through c) for multiple ones of said plurality of states.

7. An interactive voice response system having a dialogue model comprising:

a plurality of states and a plurality of interconnecting edges;

a current state;

user dialogue data indicating for a user a total number of visits to a state,

wherein each state represents a stage of the dialogue and each interconnecting edge represents a transition between one stage of the dialogue and a subsequent stage,

the system further comprising;

stage repetition monitoring means arranged in operation to find the number of times the current stage has been repeated in the current and previous dialogues from said user dialogue data,

message wording selection means arranged in operation to select the wording of a message for sending to the user at said current repeated stage, said message wording being selected in dependence upon

a) the current stage in the dialogue; and

b) the number of times the current stage has been repeated; and

c) the total number of times the user has been at his stage during one or more previously dialogues; and

repeating a) through c) for multiple ones of said plurality of states.

9. A system according to claim 7, wherein the message wording selection means is further operable to select the message wording in dependence on further data indicating whether the user has earlier visited the current stage during the current dialogue and upon data indicating the message wording which was selected on the most recent occasion on which the user was at the current stage.

13. A system as in claim 7, wherein the user enters a unique identifier and a personal identification number which must match an entry in a user access database before the user can access the system.

14. A system as in claim 13, wherein the user enters the unique identifier and personal identification number using a telephone keypad.

15. A system as in claim 13, wherein the user enters the unique identifier and personal identification number by speaking into a telephone.