System and method of providing a spoken dialog interface to a website

• US 8,949,132 B2
• Filed: 03/25/2014
• Issued: 02/03/2015
• Est. Priority Date: 12/18/2002
• Status: Expired due to Fees

• Current Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Original Assignee: AT&T Intellectual Property II LP (AT&T, Inc.)
• Inventors: Bangalore, Srinivas, Feng, Junlan, Rahim, Mazin G.
• Primary Examiner(s): YEN, ERIC L
• Application Number:
  US14/224,644
• Publication Number:
  US 20140207462A1
• Time in Patent Office: 315 Days
• Field of Search: 704270-275, 704/9
• US Class Current: 704/270.1
• CPC Class Codes: G10L 13/08 Text analysis or generation of parameters for speech synthesis out of text, e.g. grapheme to phoneme translation, prosody generation or stress or intonation determination
• CPC Class Codes: G10L 15/063 Training
• CPC Class Codes: G10L 15/183 using context dependencies, e.g. language models
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: H04M 3/4936 Speech interaction details speech recognition per se G10L15/00

Abstract

Disclosed is a method for training a spoken dialog service component from website data. Spoken dialog service components typically include an automatic speech recognition module, a language understanding module, a dialog management module, a language generation module and a text-to-speech module. The method includes selecting anchor texts within a website based on a term density, weighting those anchor texts based on a percent of salient words to total words, and incorporating the weighted anchor texts into a live spoken dialog interface, the weights determining a level of incorporation into the live spoken dialog interface.

Claims

1. A method comprising:

weighting an anchor text in a website, wherein a weight of the anchor text is based on a ratio of salient words in the website to total words in the website, to yield a weighted anchor text; and

incorporating the weighted anchor text into a live spoken dialog system by using the weighted anchor text in a live spoken dialog of the live spoken dialog system, the live spoken dialog system engaging in the live spoken dialog with a user, wherein a weight of the weighted anchor text determines a level of incorporation of the weighted anchor text into the live spoken dialog.

2. The method of claim 1, further comprising extracting a linguistic item from the website.

3. The method of claim 2, further comprising generating a website specific language model using the linguistic item and the weighted anchor text.

4. The method of claim 3, further comprising integrating the website specific language model into the live spoken dialog.

5. The method of claim 2, wherein the level of incorporation is based on the linguistic item.

6. The method of claim 2, wherein the linguistic item comprises one of a named entity, a nominal phrase, a verbal phrase, and an adjectival phrase.

7. The method of claim 1, further comprising computing a top anchor text from a plurality of weighted anchor texts, wherein the plurality of weighted anchor texts comprises the weighted anchor text, the top anchor text having a highest ratio of salient words to total words from the plurality of weighted anchor texts.

8. A system comprising:

a processor; and

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

weighting an anchor text in a website, wherein a weight of the anchor text is based on a ratio of salient words in the website to total words in the website, to yield a weighted anchor text; and

incorporating the weighted anchor text into a live spoken dialog system by using the weighted anchor text in a live spoken dialog of the live spoken dialog system, the live spoken dialog system engaging in the live spoken dialog with a user, wherein a weight of the weighted anchor text determines a level of incorporation of the weighted anchor text into the live spoken dialog.

9. The system of claim 8, the computer-readable storage medium having additional instructions stored which, when executed by the processor, result in operations comprising extracting a linguistic term from the website.

10. The system of claim 9, the computer-readable storage medium having additional instructions stored which, when executed by the processor, result in operations comprising generating a website specific language model, specific to the website, using the linguistic item and the weighted anchor text.

11. The system of claim 10, the computer-readable storage medium having additional instructions stored which, when executed by the processor, result in operations comprising integrating the website specific language model into the live spoken dialog.

12. The system of claim 9, wherein the level of incorporation is based on the linguistic item.

13. The system of claim 9, wherein the linguistic item comprises one of a named entity, a nominal phrase, a verbal phrase, and an adjectival phrase.

14. The system of claim 8, the computer-readable storage medium having additional instructions stored which, when executed by the processor, result in operations comprising computing a top anchor text from a plurality of weighted anchor texts, wherein the plurality of weighted anchor texts comprises the weighted anchor text, the top anchor text having a highest ratio of salient words to total words from the plurality of weighted anchor texts.

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

weighting an anchor text in a website, wherein a weight of the anchor text is based on a ratio of salient words in the website to total words in the website, to yield a weighted anchor text; and

incorporating the weighted anchor text into a live spoken dialog system by using the weighted anchor text in a live spoken dialog of the live spoken dialog system, the live spoken dialog system engaging in the live spoken dialog with a user, wherein a weight of the weighted anchor text determines a level of incorporation of the weighted anchor text into the live spoken dialog.

16. The computer-readable storage device of claim 15, having additional instructions stored which, when executed by the computing device, result in operations further comprising extracting a linguistic term from the website.

17. The computer-readable storage device of claim 16, having additional instructions stored which, when executed by the computing device, result in operations further comprising generating a website specific language model, specific to the website, using the linguistic item and the weighted anchor text.

18. The computer-readable storage device of claim 17, having additional instructions stored which, when executed by the computing device, result in operations further comprising integrating the website specific language model into the live spoken dialog.

19. The computer-readable storage device of claim 16, wherein the level of incorporation is based on the linguistic item.

20. The computer-readable storage device of claim 16, wherein the linguistic item comprises one of a named entity, a nominal phrase, a verbal phrase, and an adjectival phrase.