METHOD AND APPARATUS FOR PROCESSING DATA

• US 20110046943A1
• Filed: 04/05/2010
• Published: 02/24/2011
• Est. Priority Date: 08/19/2009
• Status: Active Grant

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Samsung Electronics Co. Ltd.
• Inventors: Ahn, Jae Hyun, Lee, Dong Yeol, Park, Seung Seop
• Granted Patent: US 8,626,489 B2
• Time in Patent Office: Days
• US Class Current: 704/9
• CPC Class Codes: G06F 40/20 Natural language analysis semantic analysis of natural language G06F40/30
• CPC Class Codes: G10L 13/04 Details of speech synthesis systems, e.g. synthesiser structure or memory management

Abstract

A data processing method and apparatus that may set emotion based on development of a story are provided. The method and apparatus may set emotion without inputting emotion for each sentence of text data. Emotion setting information is generated based on development of the story and the like, and may be applied to the text data.

Claims

1. A method for processing text data, the method comprising:

receiving text data comprising a story;

generating a profile with respect to at least one of a character in the story and a narrator in the story; and

generating emotion information corresponding to the story, the emotion information relating to at least one of a profile with respect to a character in the story, a profile with respect to a narrator in the story, emotion change information of a character in the story, and tension information of the story.

2. The method of claim 1, wherein the profile with respect to the character comprises at least one of a character name, a characteristic of the character, a speech style, a relationship with another character, a speed of speech, an intensity of speech, and a pitch of speech, and the profile with respect to the narrator comprises at least one of a narrator name, a speech style, a degree of applying a tension of the story, a speed of speech, an intensity of speech, and a pitch of speech.

3. The method of claim 1, wherein the generating of the emotion information comprises:

dividing the story based on a classifying standard;

generating speaker information with respect to the text data based on the divided story; and

generating at least one of conversation background information corresponding to each division of the story, emotion change information of the character, and tension information of the story.

4. The method of claim 3, wherein the classifying standard is chapter information included in the text data or a classifying standard inputted by a user.

5. The method of claim 3, wherein the generating of the speaker information comprises:

analyzing a sentence included in the text data; and

generating speaker information based on the analyzed sentence.

6. The method of claim 1, further comprising mapping emotion information corresponding to the story to the text data and storing the mapped information.

7. An apparatus for storing text data, the apparatus comprising:

a text data storing unit configured to store text data comprising a story;

a profile storing unit configured to store a profile with respect to at least one of a character in the story and a narrator in the story; and

an emotion information storing unit configured to store emotion information corresponding to the story, the emotion information relating to at least one of a profile with respect to a character in the story, a profile with respect to a narrator in the story, emotion change information of a character in the story, and tension information of the story.

8. The apparatus of claim 7, wherein the profile with respect to the character comprises at least one of a character name, a characteristic of the character, a speech style, a relationship with another character, a speed of speech, an intensity of speech, and a pitch of speech, and the profile with respect to the narrator comprises at least one of a narrator name, a speech style, a degree of applying a tension of the story, a speed of speech, an intensity of speech, and a pitch of speech.

9. The apparatus of claim 7, wherein the text data comprising the story is divided based on a classifying standard, and the emotion information storing unit stores at least one of conversation background information corresponding to each division of the text data, emotion change information of a character, and tension information of the story.

10. The apparatus of claim 9, wherein the classifying standard is at least one of chapter information included in the text data, a change in a situation where the story is developed, a conversation background in the story, an emotion change of a character, a change in scene of the story, and depth information set to divide the story.

11. A method for processing text data, the method comprising:

loading text data comprising a story;

reading emotion information corresponding to the story; and

generating speech setting information to perform a text-to-speech (TTS) transformation based on the emotion information, the emotion information relating to at least one of a profile with respect to a character in the story, a profile with respect to a narrator in the story, emotion change information of a character in the story, tension information of the story, and conversation background information in the story.

12. The method of claim 11, wherein, when the speaker is a narrator, the reading of the emotion information comprises:

reading the emotion information based on a sentence unit of the text data or a speaker unit; and

extracting, from the emotion information, tension information of a current point in time.

13. The method of claim 11, wherein, when the character is a speaker, the reading of the emotion information comprises:

reading the emotion information based on a sentence unit of the text data or a speaker unit; and

extracting, from the emotion change information of the character, emotion information of the character of a current point in time.

14. The method of claim 11, wherein the generating of the speech setting information comprises:

setting background music information corresponding to tension information of the story; and

setting acoustic effect information corresponding to conversation background information in the story.

15. The method of claim 11, further comprising transforming the text data into speech based on the speech setting information.

16. An apparatus for processing text data, the apparatus comprising:

a text data loading unit configured to load text data comprising a story;

an emotion information reading unit to read emotion information corresponding to the story;

a speech setting information generating unit configured to generate speech setting information to perform a TTS transformation based on the emotion information; and

a TTS transforming unit configured to transform the text data to speech based on the speech setting information.

17. The apparatus of claim 16, wherein the emotion information is at least one of a profile with respect to a character in the story, a profile with respect to a narrator in the story, emotion change information of a character in the story, tension information of the story, and conversation background information in the story.

18. The apparatus of claim 17, wherein the profile with respect to the character comprises at least one of a character name, a characteristic of the character, a speech style, a relationship of another character, a speed of speech, an intensity of speech, and a pitch of speech, and the profile with respect to the narrator comprises at least one of a narrator name, a speech style, a degree of applying a tension of a story, a speed of speech, an intensity of speech, and a pitch of speech.

19. The apparatus of claim 16, wherein the speech setting information comprises at least one of background music information corresponding to the tension information of the story or acoustic effect information corresponding to conversation background information in the story.

20. The apparatus of claim 16, wherein the speech setting information comprises:

extracting tension information of a current point in time based on a sentence unit of the text data or a speaker unit; and

generating speech setting information of the current point in time, from the tension information of the current point in time,

wherein, when the tension information of the current point in time does not exist, generating speech setting information of the current point in time based on tension information of a past point in time or a future point in time.

21. The apparatus of claim 16, wherein the speech setting information generating unit generates speech setting information of a current point in time from emotion change information of the character.