Method and System for Enhancing a Speech Database

• US 20140278431A1
• Filed: 05/28/2014
• Published: 09/18/2014
• Est. Priority Date: 08/31/2006
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: AT&T Intellectual Property II LP (AT&T, Inc.)
• Inventors: Conkie, Alistair D., Syrdal, Ann K.
• Granted Patent: US 8,977,552 B2
• Time in Patent Office: Days
• US Class Current: 704/260
• CPC Class Codes: G10L 13/00 Speech synthesis; Text to speech systems
• CPC Class Codes: G10L 13/02 Methods for producing synthetic speech; Speech synthesisers
• CPC Class Codes: G10L 13/06 Elementary speech units used in speech synthesisers; Concatenation rules
• 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

Abstract

A system, method and computer readable medium that enhances a speech database for speech synthesis is disclosed. The method may include labeling audio files in a primary speech database, identifying segments in the labeled audio files that have varying pronunciations based on language differences, identifying replacement segments in a secondary speech database, enhancing the primary speech database by substituting the identified secondary speech database segments for the corresponding identified segments in the primary speech database, and storing the enhanced primary speech database for use in speech synthesis.

Claims

1. A method comprising:

selecting, via a processor, a speech segment associated with text, wherein the speech segment is selected from a primary speech database which has been modified by;

identifying primary speech segments in the primary speech database which do not meet a need of a text-to-speech process;

identifying replacement speech segments which satisfy the need in a secondary speech database; and

enhancing the primary speech database by substituting, in the primary database, the primary speech segments with the replacement speech segments; and

generating, via the processor, speech corresponding to the text using the speech segment.

2. The method of claim 1, wherein the primary speech database has been further modified by identifying boundaries of the primary speech segments.

3. The method of claim 2, wherein phone boundaries of the primary speech segments are identified using a zero-crossing calculation.

4. The method of claim 1, wherein the need is based on one of dialect differences, geographic language differences, regional language differences, accent differences, national language differences, idiosyncratic speech differences, and database coverage differences.

5. The method of claim 1, wherein the primary speech segments are one of diphones, triphones, and phonemes.

6. The method of claim 1, wherein the primary speech database comprises first voice recordings in a first dialect, and the secondary speech database comprises second voice recordings in a second dialect, wherein the first dialect and the second dialect differ by one of dialect differences, geographic language differences, regional language differences, accent differences, national language differences, idiosyncratic speech differences, and database coverage differences.

7. The method of claim 1, wherein the primary speech segments are identified based on one of obstruents and nasals.

8. The method of claim 1, wherein phone boundaries of the primary speech segments are identified using a zero-crossing calculation.

9. 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;

selecting, via a processor, a speech segment associated with text, wherein the speech segment is selected from a primary speech database which has been modified by;

identifying primary speech segments in the primary speech database which do not meet a need of a text-to-speech process;

identifying replacement speech segments which satisfy the need in a secondary speech database; and

enhancing the primary speech database by substituting, in the primary database, the primary speech segments with the replacement speech segments; and

generating, via the processor, speech corresponding to the text using the speech segment.

10. The system of claim 9, wherein the primary speech database has been further modified by identifying boundaries of the primary speech segments.

11. The system of claim 10, wherein phone boundaries of the primary speech segments are identified using a zero-crossing calculation.

12. The system of claim 9, wherein the need is based on one of dialect differences, geographic language differences, regional language differences, accent differences, national language differences, idiosyncratic speech differences, and database coverage differences.

13. The system of claim 9, wherein the primary speech segments are one of diphones, triphones, and phonemes.

14. The system of claim 9, wherein the primary speech database comprises first voice recordings in a first dialect, and the secondary speech database comprises second voice recordings in a second dialect, wherein the first dialect and the second dialect differ by one of dialect differences, geographic language differences, regional language differences, accent differences, national language differences, idiosyncratic speech differences, and database coverage differences.

15. The system of claim 9, wherein the primary speech segments are identified based on one of obstruents and nasals.

16. The system of claim 9, wherein phone boundaries of the primary speech segments are identified using a zero-crossing calculation.

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

selecting, via a processor, a speech segment associated with text, wherein the speech segment is selected from a primary speech database which has been modified by;

identifying primary speech segments in the primary speech database which do not meet a need of a text-to-speech process;

identifying replacement speech segments which satisfy the need in a secondary speech database; and

enhancing the primary speech database by substituting, in the primary database, the primary speech segments with the replacement speech segments; and

generating, via the processor, speech corresponding to the text using the speech segment.

18. The computer-readable storage device of claim 17, wherein the primary speech database has been further modified by identifying boundaries of the primary speech segments.

19. The computer-readable storage device of claim 17, wherein phone boundaries of the primary speech segments are identified using a zero-crossing calculation.

20. The computer-readable storage device of claim 17, wherein the need is based on one of dialect differences, geographic language differences, regional language differences, accent differences, national language differences, idiosyncratic speech differences, and database coverage differences.