Information processing apparatus and text-to-speech method

• US 8,719,028 B2
• Filed: 12/17/2009
• Issued: 05/06/2014
• Est. Priority Date: 01/08/2009
• Status: Active Grant

• Current Assignee: Alpine Electronics Incorporated (Alps Alpine Co., Ltd.)
• Original Assignee: Alpine Electronics Incorporated (Alps Alpine Co., Ltd.)
• Inventors: Otani, Michiaki
• Primary Examiner(s): AZAD, ABUL K
• Application Number:
  US12/640,534
• Publication Number:
  US 20100174545A1
• Time in Patent Office: 1,601 Days
• Field of Search: 704258-2701, 704/275
• US Class Current: 704/260
• CPC Class Codes: G10L 13/00 Speech synthesis; Text to speech systems
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence

Abstract

An information processing apparatus for playing back data includes an oral reading unit, a storage unit storing text templates for responses to questions from a user and text template conversion rules, an input unit for inputting a question from a user, and a control unit for retrieving data and items of information associated with the data. The control unit analyzes a question about a data from a user, for example, a question about a tune, to select a text template for a response to the question and detects the characters in items of tune information of the tune. The characters are designated to replace replacement symbols included in the text template. The control unit also converts the text template based on whether the characters can be read aloud, generates a text to be read aloud using the converted text template, and causes the oral reading unit to read the text aloud.

Claims

1. An information processing apparatus in communication with an apparatus configured to store and playback audio data, the audio data including tune data, and tune information in the form of character text or word text corresponding to the tune data, the information processing apparatus comprising:

an oral reading unit configured to convert text to speech data, and output the speech data;

an input unit for inputting a question from a user corresponding to the tune data;

a storage unit configured to store a plurality of text templates corresponding to predetermined responses to the questions, and configured to store a plurality of text template conversion rules, each text template having a predetermined associated genre;

a control unit configured to retrieve the tune information associated with the tune data;

wherein the control unit analyzes the question from the user corresponding to the tune data, and selects a text template corresponding to a response to the question and based upon the predetermined associated genre, the selected text template having substitution symbols corresponding to the character text or word text associated with the tune data;

wherein if the control unit determines that the substitution symbols in the selected text template corresponding to the character text or word text cannot be read aloud, a new text template is generated based on the selected text template, and the substitution symbols that cannot be read aloud are modified in accordance with the conversion rules such that all portions of the new text template can be read aloud, and wherein the new text template read aloud contains no silent portions; and

wherein the control unit causes the oral reading unit to read the text aloud corresponding to the new text template.

2. The information processing apparatus according to claim 1, wherein the control unit selects the text template based on the genre of a tune corresponding to the question.

3. The information processing apparatus according to claim 1, wherein if at least one of the substitution symbols cannot be read aloud, the control unit generates modified text data using the selected text template, wherein the substitution symbol is removed or replaced.

4. The information processing apparatus according to claim 3, wherein if a substitution symbol corresponding to an artist name cannot be read aloud, the control unit generates the new text template such that the substitution symbol corresponding to the artist name, is removed.

5. The information processing apparatus according to claim 3, wherein if a substitution symbol corresponding to an album name cannot be read aloud, the control unit generates the new text template such that the substitution symbol corresponding to the album name, is removed.

6. The information processing apparatus according to claim 3, wherein if a substitution symbol corresponding to a tune name cannot be read aloud, the control unit extracts, from the storage unit, a new text template having a substitution symbol corresponding to an artist name of the tune as a first component, and having a substitution symbol corresponding to an album name of the tune as a second component, and generates the new text template.

7. The information processing apparatus according to claim 3, wherein if only a substitution symbol corresponding to an artist name can be read aloud, the control unit extracts, from the storage unit, a new text template having a substitution symbol corresponding to the artist name as a component, and generates the new text template.

8. The information processing apparatus according to claim 3, wherein if only a substitution symbol corresponding to an album name can be read aloud, the control unit extracts, from the storage unit, a new text template having a substitution symbol corresponding to the album name as a component and generates the new text template.

9. The information processing apparatus according to claim 3, wherein if only a substitution symbol corresponding to a tune name can be read aloud, the control unit generates the new text template such that the substitution symbols corresponding to an artist name of the tune and an album name of the tune, are removed.

10. A text-to-speech method comprising the steps of:

a) analyzing a question received by an information processing apparatus from a user regarding audio data stored in a storage unit, the audio data including tune data, and tune information in the form of character text or word text corresponding to the tune data, wherein the information processing apparatus is configured to convert text data to speech data;

b) extracting a text template from a plurality of predetermined text templates stored in the storage unit, the extracted text template corresponding to the question analyzed and corresponding to a predetermined associated genre;

c) detecting substitution symbols in the extracted text template corresponding to the character text or word text associated with the tune data;

d) converting the extracted text template to a new text template if the substitution symbols in the extracted text template corresponding to the character text or word text cannot be read aloud;

e) replacing the substitution symbols that cannot be read aloud with substitution symbols that can be read aloud, or removing the substitution symbols that cannot be read aloud, in accordance with predetermined conversion rules; and

f) converting the text of the new text template to speech data for output of the speech data, and wherein the new text template read aloud contains no silent portions.

11. The text-to-speech method according to claim 10, further comprising:

selecting candidates for the text template from the plurality of text templates stored in the storage unit, wherein the candidates are selected that provide a response to the question;

detecting the genre corresponding to the tune data that is a subject of the question, by analyzing the tune information corresponding to the tune data, and

selecting a text template corresponding to the detected genre from the candidates.

12. The text-to-speech method according to claim 11, further comprising:

g) determining if at least one of the substitution symbols cannot be read aloud; and

h) converting the text template to the new text template by removing or replacing the substitution symbols that cannot be read aloud.

13. The text-to-speech method according to claim 12, wherein if a substitution symbol corresponding to an artist name cannot be read aloud, the information processing apparatus converts the text template to the new text template such that the substitution symbol corresponding to the artist name is removed.

14. The text-to-speech method according to claim 12, wherein if a substitution symbol corresponding to an album name cannot be read aloud, the information processing apparatus converts the text template to the new text template such that the substitution symbol corresponding to the album name is removed.

15. The text-to-speech method according to claim 12, wherein if a substitution symbol corresponding to a tune name cannot be read aloud, the information processing apparatus generates the new text template from the storage unit having a substitution symbol corresponding to an artist name of the tune as first component, and having a substitution symbol corresponding to an album name of the tune as a second component, to generate the new text template.

16. The text-to-speech method according to claim 12, wherein if only a substitution symbol corresponding to an artist name can be read aloud, the information processing apparatus extracts the new text template having a substitution symbol corresponding to the artist name as a component.

17. The text-to-speech method according to claim 12, wherein if only a substitution symbol corresponding to an album name can be read aloud, the information processing apparatus extracts a new text template having a substitution symbol corresponding to the album name as a component.

18. The text-to-speech method according to claim 12, wherein if only a substitution symbol corresponding to a tune name can be read aloud, the information processing apparatus generates the text template such that the substitution symbols corresponding to an artist name of the tune, and an album name of the tune, are removed.