Method of handling frequently asked questions in a natural language dialog service

• US 8,645,122 B1
• Filed: 12/19/2002
• Issued: 02/04/2014
• Est. Priority Date: 12/19/2002
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: AT&T Intellectual Property II LP (AT&T, Inc.)
• Inventors: Dutton, Dawn L, Hollister, Barbara B., Gupta, Narendra K., Rahim, Mazin G, Riccardi, Giuseppe, Schapire, Robert Elias, Schroeter, Juergen, Di Fabbrizio, Giuseppe
• Primary Examiner(s): Spooner, Lamont
• Application Number:
  US10/325,266
• Time in Patent Office: 4,065 Days
• Field of Search: 704/9, 704/231, 704/251, 704/257, 707 2- 5
• US Class Current: 704/9
• CPC Class Codes: G06F 40/237 Lexical tools
• CPC Class Codes: G10L 13/02 Methods for producing synthetic speech; Speech synthesisers
• CPC Class Codes: G10L 21/00 Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility G10L19/00 takes precedence

Abstract

A voice-enabled help desk service is disclosed. The service comprises an automatic speech recognition module for recognizing speech from a user, a spoken language understanding module for understanding the output from the automatic speech recognition module, a dialog management module for generating a response to speech from the user, a natural voices text-to-speech synthesis module for synthesizing speech to generate the response to the user, and a frequently asked questions module. The frequently asked questions module handles frequently asked questions from the user by changing voices and providing predetermined prompts to answer frequently asked questions.

Claims

1. A method comprising:

receiving an utterance, in a turn-based dialog, from a user;

selecting, via a processor, a prompt type to convey information to the user in response to the utterance, to yield a selected prompt type;

developing, via a processor, a back story for each pre-synthesized voice in a plurality of pre-synthesized voices;

retrieving, based on the selected prompt type, a pre-synthesized voice from the plurality of pre-synthesized voices, where the prompt type is associated with a specific pre-synthesized voice in the plurality of pre-synthesized voices;

generating a prompt, as part of the turn-based dialog, by inputting the back story, a dialog history, the selected prompt type, and a set of logical predicates to an interpreter tuple; and

replying to the utterance in the turn-based dialog via the prompt using the pre-synthesized voice.

2. The method of claim 1, wherein the back story is developed based on a set of personality traits associated with each pre-synthesized voice in the plurality of pre-synthesized voices.

3. The method of claim 1, wherein the prompt type is one of a pricing, a hint, a demonstration, a question, an example, an instruction, and a command.

4. The method of claim 1, wherein selecting the prompt type further comprises:

performing an analysis to determine which prompt type of a plurality of prompt types is most likely to be understood by the turn-based dialog, to yield a highest probability prompt type; and

selecting the highest probability prompt type as the prompt type.

5. The method of claim 1, wherein there are eight pre-synthesized voices in the plurality of pre-synthesized voices.

6. A system, comprising:

a processor; and

a computer-readable storage device having instructions stored which, when executed by the processor, perform operations comprising;

receiving an utterance, in a turn-based dialog, from a user;

selecting a prompt type to convey information to the user in response to the utterance, to yield a selected prompt type;

developing, via the processor, a back story for each pre-synthesized voice in a plurality of pre-synthesized voices;

retrieving, based on the selected prompt type, a pre-synthesized voice from the plurality of pre-synthesized voices, where the prompt type is associated with a specific pre-synthesized voice in the plurality of pre-synthesized voices; and

generating a prompt, as part of the turn-based dialog, by inputting the back story, a dialog history, the selected prompt type, and a set of logical predicates to an interpreter tuple; and

replying to the utterance in the turn-based dialog via the prompt using the pre-synthesized voice.

7. The system of claim 6, wherein the back story is developed based on a set of personality traits associated with each pre-synthesized voice in the plurality of pre-synthesized voices.

8. The system of claim 6, wherein the prompt type comprises one of a pricing, a hint, a demonstration, a question, an example, an instruction, and a command.

9. The system of claim 8, wherein there are eight pre-synthesized voices in the plurality of pre-synthesized voices.

10. The system of claim 6, wherein selecting the prompt type further comprises:

performing an analysis to determine which prompt type in a plurality of prompt types is most likely to be understood by the turn-based dialog, to yield a highest probability prompt type; and

selecting the highest probability prompt type as the prompt type.

11. A computer-readable storage device having instructions stored which, when executed by a computing device, cause the computing device to perform operations comprising:

receiving an utterance, in a turn-based dialog, from a user;

selecting a prompt type to convey information to the user in response to the utterance, to yield a selected prompt type;

developing, via the computing device, a back story for each pre-synthesized voice in a plurality of pre-synthesized voices;

retrieving, based on the selected prompt type, a pre-synthesized voice from the plurality of pre-synthesized voices, where the prompt type is associated with a specific pre-synthesized voice in the plurality of pre-synthesized voices; and

generating a prompt, as part of the turn-based dialog, by inputting the back story, a dialog history, the selected prompt type, and a set of logical predicates to an interpreter tuple; and

replying to the utterance in the turn-based dialog via the prompt using the pre-synthesized voice.

12. The computer-readable storage device of claim 11, wherein the back story is developed based on a set of personality traits associated with each pre-synthesized voice in the plurality of pre-synthesized voices.

13. The computer-readable storage device of claim 11, wherein the prompt type comprises one of a pricing, a hint, a demonstration, a question, an example, an instruction, and a command.

14. The computer-readable storage device of claim 11, wherein selecting the prompt type further comprises:

performing an analysis to determine which prompt type of a plurality of prompt types is most likely to be understood by the turn-based dialog, to yield a highest probability prompt type; and

selecting the highest probability prompt type as the prompt type.

15. The computer-readable storage device of claim 11, wherein there are eight pre-synthesized voices in the plurality of pre-synthesized voices.