System and method for providing information using spoken dialogue interface

• US 7,225,128 B2
• Filed: 03/31/2003
• Issued: 05/29/2007
• Est. Priority Date: 03/29/2002
• Status: Active Grant

• Current Assignee: Samsung Electronics Co. Ltd. (Samsung Group)
• Original Assignee: Samsung Electronics Co. Ltd. (Samsung Group)
• Inventors: Lee, Jae-won, Lee, Hye-jeong, Lee, Yong-beom, Park, Chan-min, Kim, Jeong-su, Seo, Hee-kyoung
• Primary Examiner(s): CHAWAN, VIJAY B
• Application Number:
  US10/401,695
• Publication Number:
  US 20030220799A1
• Time in Patent Office: 1,520 Days
• Field of Search: 704/9, 704/275, 704/257, 704/260, 704/235, 704/258, 704/270.1, 704/1
• US Class Current: 704/257
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue

Abstract

There are provided a system and method for providing information using a spoken dialogue interface. The system includes a speech recognizer for transforming voice signals into sentences; a sentence analyzer for analyzing the sentences by their structural elements; a dialogue manager for extracting information on speech acts or intentions from the structural elements, and generating information on system'"'"'s speech acts or intentions for a response to the extracted information on speech acts or intentions; a sentence generator for generating sentences based on the information on the system'"'"'s speech acts or intentions for the response; a speech synthesizer for synthesizing the generated sentences into voices; an information extractor for extracting information required for the response from the Internet in real time; and a user modeling means for analyzing and classifying users'"'"' tendencies. Information demanded by a user can be detected in real time and provided through a voice interface with versatile and familiar dialogues based on the user'"'"'s tendencies.

Claims

1. A system for providing information using a spoken dialogue interface, comprising:

a speech recognizer for transforming voice signals into sentences;

a sentence analyzer for analyzing the sentences by their meaning structures;

a dialogue manager for extracting information on user speech acts or intentions from the meaning structures and generating information on system speech acts or intentions for a response to the extracted information on the user'"'"'s speech acts or intentions;

a sentence generator for generating sentences based on the information on the system'"'"'s speech acts or intentions for the response to the voice signals; and

a speech synthesizer for synthesizing the generated sentences into voices.

2. The system of claim 1, wherein the sentence analyzer includes a morphological analyzer for separating the sentences into their morphemes and tagging the separated morphemes and a syntactic analyzer for analyzing sentence structural elements based on relationships between the morphemes.

3. The system of claim 1, wherein the sentence analyzer further includes a semantic analyzer for transforming the sentences transformed from the voice signals into the meaning structures.

4. The system of claim 1, wherein the dialogue manager includes an intention analyzer for determining the user'"'"'s speech acts or intentions from the meaning structures, and an intention generator for generating the system'"'"'s speech acts or intentions for a response to the user'"'"'s speech acts or intentions.

5. The system of claim 1, further comprising a query generator for generating query information based on information on the user'"'"'s speech acts or intentions.

6. The system of claim 5, further comprising an information extractor for extracting information using the query information as key words and a user modeling means for modeling the user'"'"'s tendencies from the user'"'"'s dialogues.

7. The system of claim 1, further comprising a knowledge database for storing the information on the user'"'"'s speech acts or intentions extracted from the meaning structures, and the information on the system'"'"'s speech acts or intentions.

8. The system of claim 1, wherein the sentence generator includes a sentence structure generator for receiving the information on the system'"'"'s speech acts or intentions and generating sentence structures and a morpheme generator for receiving the sentence structures and generating morphemes.

9. A method of providing information using a spoken dialogue interface, comprising the steps of:

(a) transforming voice signals into sentences;

(b) analyzing the sentences by their meaning structures;

(c) extracting information on user speech acts or intentions from the meaning structures, and generating information on system speech acts or intentions for a response to the extracted information on the user'"'"'s speech acts or intentions;

(d) generating sentences based on the information on the system'"'"'s speech acts or intentions for the response to the voice signals; and

(e) synthesizing the generated sentences into voices.

10. The method for claim 9, wherein step (b) includes (b1) separating the sentences into their morphemes and tagging the separated morphemes and (b2) analyzing structural elements of a sentence based on relationships between the morphemes.

11. The method of claim 9, wherein the step (c) includes (c1) determining the user'"'"'s speech acts or intentions from the meaning structures, (c2) searching through a dialogue case database based on information on the user'"'"'s speech acts or intentions, (c3) calculating similarities of the detected dialogue cases using information on the user'"'"'s speech acts or intentions and information on the user'"'"'s tendencies, (c4) selecting the most similar dialogue case using information on the similarities and determining the system'"'"'s speech acts or intentions for a system response, (c5) generating query information for a response, and (c6) receiving search results obtained through the query information and completing the system'"'"'s speech acts or intentions.

12. The method of claim 9, wherein the step (b) includes transforming the sentences transformed from the voice signals into the meaning structures.

13. The method of claim 9, wherein the step (c) includes storing the information on the user'"'"'s speech acts or intentions extracted from the meaning structures.

14. The method of claim 9, wherein step (d) includes (d1) generating sentence structures based on the information on the system'"'"'s speech acts or intentions and (d2) generating morphemes for a response.

15. A computer readable recording medium that stores a program for the computer to implement the method claimed in claim 9.