Interactive dialogues

• US 7,143,040 B2
• Filed: 07/19/2001
• Issued: 11/28/2006
• Est. Priority Date: 07/20/2000
• Status: Expired due to Fees

• Current Assignee: RPX Corporation
• Original Assignee: British Telecommunications PLC (BT Group PLC)
• Inventors: Attwater, David J, Durston, Peter J
• Primary Examiner(s): MCFADDEN, SUSAN IRIS
• Application Number:
  US10/312,984
• Publication Number:
  US 20030130849A1
• Time in Patent Office: 1,958 Days
• Field of Search: None
• US Class Current: 704/270
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue

Abstract

An interactive dialogue apparatus generates prompts and receives responses which it processes and stores response data. Progress of the dialogue is controlled using a state model which is parsed, comparing transition conditions with the stored response data, stopping at a state for which no edge condition is satisfied. Each time a fresh response is received, the model is reparsed from the start. Prompts are generated by reference to a state history, being selected by finding the longest match between the history and state sequences associated with stored prompts. Prompts are compared so that the dialogue may be terminated in the event of repetition.

Claims

1. An interactive dialogue apparatus comprising:

means for generating prompts to be output;

means for processing input responses to produce response data;

a store for storing response data;

a store storing data defining an initial state, a plurality of subsequent states, possible transitions between states, and for each transition an associated condition to be satisfied before that transition is deemed allowable, at least some of said conditions involving the stored response data; and

control means operable to

(a) perform a parsing operation in which it examines whether transition conditions are satisfied so as to identify a subsequent state in respect of which—

(i) the condition(s) associated with the transition(s) from a reference state to that state are satisfied, and

(ii) no conditions associated with any transition from that state are satisfied;

(b) select, in dependence upon the state thus identified, a prompt to be output;

wherein the control means is operable upon receipt of a response to repeat the parsing operation using said initial state as the reference state.

2. An apparatus according to claim 1, including a state history store, the control means being operable to record the identified state in the history store;

wherein the apparatus includes a prompt definition store for storing a plurality of prompt definitions, each prompt definition having a sequence of one or more states associated with it, and the control means is arranged to select the prompt in dependence upon the longest sequence of states in the prompt definition store which sequence of states is equal to a sequence of the most recent one or more states stored in the state history.

3. An apparatus according to claim 1 including means for storing the most recent prompt that has been output;

and means operable to compare the selected prompt with the stored most recent prompt and in the event of a match to terminate the dialogue process.

4. An apparatus according to claim 1, in which the or a store for storing a plurality of prompt messages contains data representing for each message an associated cost and in which the control means is operable, when the message is generated, to add the associated cost to a total cost, and, if the total cost exceeds a predetermined monitor value, to connect the apparatus to a monitoring station.

5. An apparatus according to claim 1, in which the or a store for storing a plurality of prompt messages contains data representing for each message an associated cost and in which the control means is operable when the message is generated, to add the associated cost to a total cost, and, if the total cost exceeds a predetermined transfer value, to transfer operation to an operator station and disconnect the apparatus.

6. An apparatus according to claim 1 for use with spoken responses, in which the response processing means includes a speech recogniser.

7. An apparatus according to claim 1 for use with spoken prompts, in which the prompt generating means includes a speech synthesiser.

8. An interactive dialogue apparatus comprising:

means for generating prompts to be output;

means for processing input responses to produce response data;

a store for storing response data;

a store storing data defining an initial state, a plurality of subsequent states, possible transitions between states, and for each transition an associated condition to be satisfied before that transition is deemed allowable, at least some of said conditions involving the stored response data;

a state history store; and

control means operable to

(a) perform a parsing operation in which it examines whether transition conditions are satisfied so as to identify a subsequent state in respect of which—

(i) the condition(s) associated with the transition(s) from a reference state to that state are satisfied, and

(ii) no conditions associated with any transition from that state are satisfied;

(b) record the identified state in the history store;

(c) select, in dependence upon the contents of the history store, a prompt to be output;

wherein the apparatus includes a prompt definition store for storing a plurality of prompt definitions, each prompt definition having a sequential pattern of one or more states associated with it, and the control means is arranged to select, of those prompts having associated with them a sequential pattern which matches at least the most recent part of the sequence of states in the state history store, the prompt having the longest such match.

9. An interactive dialogue apparatus comprising:

means for generating prompts to be output;

means for processing input responses to produce response data;

control means operable to perform a dialogue process in which it selects, in dependence upon the response data, a prompt to be output;

means for storing the most recent prompt that has been output;

means operable to compare the selected prompt with the stored most recent prompt and in the event of a match to terminate the dialogue process.

10. An apparatus according to claim 9 in which the control means is operable to store the prompts in text form, and to perform said comparisons upon text strings.

11. An apparatus according to claim 9 in which the control means is operable to store the prompts in the form of a list of one or more file names representing stored speech, and to perform said comparisons upon file name lists.

12. A method of operating an interactive dialogue with a user comprising:

processing input responses to produce response data;

storing response data;

performing a parsing operation by reference to a dialogue model defined by stored data defining an initial state, a plurality of subsequent states, possible transitions between states, and for each transition an associated condition to be satisfied before that transition is deemed allowable, at least some of said conditions involving the stored response data;

wherein said parsing operation involves

(a) examining whether transition conditions are satisfied, so as to identify a subsequent state in respect of which—

(i) the condition(s) associated with the transition(s) from a reference state to that state are satisfied, and

(ii) no conditions associated with any transition from that state are satisfied;

(b) selecting, in dependence upon the state thus identified, a prompt to be output;

and wherein upon receipt of a response the parsing operation is repeated using said initial state as the reference state.

13. A method according to claim 12 including recording the identified state in the history store;

wherein the selection of the prompt is carried out by reference to a plurality of stored prompt definitions, each prompt definition having a sequence of one or more states associated with it, and selecting the prompt in dependence upon the longest sequence of states in the prompt definition store which sequence of states is equal to a sequence of the most recent one or more states stored in the state history store.

14. A method according to claim 12 including storing the most recent prompt that has been output;

and comparing the selected prompt with the stored most recent prompt and in the event of a match to terminate the dialogue process.

15. A method according to any one of claims 12 in which the responses are received in spoken form and processed by a speech recogniser.

17. A method according to claim 12, including the generation of said stored model data by

(i) generating an initial set of data;

(ii) replacing data defining a transition to a state corresponding to an implicit confirmation with data representing an additional state corresponding an explicit confirmation, a transition to said additional state, and a transition from said additional state.

18. A method according to claim 12 including the step of adding, for a transition leading from a state which corresponds to confirmation, data defining a condition that a confidence associated with the value confirmed exceeds a predetermined amount.

19. A method according to claim 12 including the step of adding, for a transition leading to a state which corresponds to confirmation, data defining a condition that a confidence associated with the value to be confirmed exceeds a predetermined amount.

20. A data carrier loadable into a computer and carrying instructions for enabling the computer to perform a method according to claims 12.

16. A method according to any one of claims including 12 including generating the prompts in spoken form.

21. A method of operating an interactive dialogue with a user comprising:

generating prompts to be output;

processing input responses to produce response data;

storing response data;

performing a parsing operation by reference to a dialogue model defined by stored data defining an initial state, a plurality of subsequent states, possible transitions between states, and for each transition an associated condition to be satisfied before that transition is deemed allowable, at least some of said conditions involving the stored response data;

wherein said parsing operation involves

(a) examining whether transition conditions are satisfied so as to identify a subsequent state in respect of which—

(i) the condition(s) associated with the transition(s) from a reference state to that state are satisfied, and

(ii) no conditions associated with any transition from that state are satisfied;

(b) recording the identified state in a history store;

(c) selecting, in dependence upon the contents of the history store, a prompt to be output;

and wherein the selection of the prompt is carried out by reference to a plurality of stored prompt definitions, each prompt definition having a sequence of one or more states associated with it, and selecting the prompt in dependence upon the longest sequence of states in the prompt definition store which sequence of states is equal to a sequence of the most recent one or more states stored in the state history store.

22. A method of operating an interactive dialogue with a user comprising:

generating prompts to be output;

processing input responses to produce response data;

performing a dialogue process involving selecting, in dependence upon the response data, a prompt to be output;

storing the most recent prompt that has been output;

comparing the selected prompt with the stored most recent prompt and in the event of a match to terminate the dialogue process.

23. A method according to claim 22 in which the prompts are stored in text form, and said comparisons are performed upon text strings.

24. A method according to claim 22 in which the prompts are stored in the form of a list of one or more file names representing stored speech, and said comparisons are performed upon file name lists.