Methods and Apparatus for Rapid Acoustic Unit Selection From a Large Speech Corpus

• US 20100286986A1
• Filed: 07/20/2010
• Published: 11/11/2010
• Est. Priority Date: 04/30/1999
• Status: Active Grant

• Current Assignee: Cerence Operating Company (Cerence Inc.)
• Original Assignee: AT&T Intellectual Property I LP (AT&T, Inc.)
• Inventors: Riley, Michael Dennis, Beutnagel, Mark Charles, Mohri, Mehryar
• Granted Patent: US 8,086,456 B2
• Time in Patent Office: Days
• US Class Current: 704/258
• CPC Class Codes: G10L 13/00 Speech synthesis; Text to speech systems
• CPC Class Codes: G10L 13/027 Concept to speech synthesisers; Generation of natural phrases from machine-based concepts generation of parameters for speech synthesis out of text G10L13/08
• CPC Class Codes: G10L 13/07 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 speech synthesis system can select recorded speech fragments, or acoustic units, from a very large database of acoustic units to produce artificial speech. The selected acoustic units are chosen to minimize a combination of target and concatenation costs for a given sentence. However, as concatenation costs, which are measures of the mismatch between sequential pairs of acoustic units, are expensive to compute, processing can be greatly reduced by pre-computing and aching the concatenation costs. Unfortunately, the number of possible sequential pairs of acoustic units makes such caching prohibitive. However, statistical experiments reveal that while about 85% of the acoustic units are typically used in common speech, less than 1% of the possible sequential pairs of acoustic units occur in practice. A method for constructing an efficient concatenation cost database is provided by synthesizing a large body of speech, identifying the acoustic unit sequential pairs generated and their respective concatenation costs, and storing those concatenation costs likely to occur. By constructing a concatenation cost database in this fraction, the processing power required at run-time is greatly reduced with negligible effect on speech quality.

Claims

1. A method comprising:

determining whether an acoustic unit sequential pair to be used for synthesizing speech has a concatenation cost in a concatenation cost database; and

if the concatenation cost database does not contain the concatenation cost for the acoustic unit sequential pair, then assigning a default value to the concatenation cost.

2. The method of claim 1, further comprising synthesizing the speech using the default value as assigned for the acoustic unit sequential pair.

3. The method of claim 1, wherein the concatenation cost database is generated by synthesizing a body of speech and identifying acoustic unit sequential pairs generated in the body of speech and their respective concatenation costs.

4. The method of claim 1, wherein the concatenation cost database contains a portion of all possible concatenation costs.

5. The method of claim 1, wherein the concatenation cost database is derived using statistical techniques which predict which acoustic unit sequential pairs are most likely to occur in common speech.

6. The method of claim 1, wherein the concatenation cost comprises a weighted sum of subcosts across phones.

7. The method of claim 1, wherein the concatenation cost provides an estimate of an acoustic mismatch between units in the acoustic unit sequential pair.

8. A system comprising:

a processor;

a first module configured to control the processor to determine whether an acoustic sequential pair to be used for synthesizing speech has a concatenation cost and a concatenation database; and

a second module configured to control the processor, if the concatenation cost database does not contain the concatenation cost for the acoustic unit sequential pair, to then assign a default value to the concatenation cost.

9. The system of claim 8, further comprising a third module configured to control the processor to synthesize the speech using the default value as assigned for the acoustic unit sequential pair.

10. The system of claim 8, wherein the concatenation cost database is generated by synthesizing a body of speech and identifying acoustic unit sequential pairs generated in the body of speech and their respective concatenation costs.

11. The system of claim 8, wherein the concatenation cost database contains a portion of all possible concatenation costs.

12. The system of claim 8, wherein the concatenation cost database is derived using statistical techniques which predict which acoustic unit sequential pairs are most likely to occur in common speech.

13. The system of claim 8, wherein the concatenation cost comprises a weighted sum of subcosts across phones.

14. The system of claim 8, wherein the concatenation cost provides an estimate of an acoustic mismatch between units in the acoustic unit sequential pair.

15. A method comprising:

determining whether an acoustic unit sequential pair to be used for synthesizing speech has a concatenation cost and a concatenation cost database; and

if the concatenation cost database does not contain the concatenation cost for the acoustic unit sequential pair, then deriving an actual concatenation cost for the acoustic unit sequential pair.

16. The method of claim 15, further comprising synthesizing the speech using the actual concatenation cost for the acoustic unit sequential pair.

17. The method of claim 15, wherein the concatenation cost database is generated by synthesizing a body of speech and identifying acoustic unit sequential pairs generated in the body of speech and their respective concatenation costs.

18. The method of claim 15, wherein the concatenation cost database contains a portion of all possible concatenation costs.

19. The method of claim 15, wherein the concatenation cost database is derived using statistical techniques which predict which acoustic unit sequential pairs are most likely to occur in common speech.

20. The method of claim 15, wherein the concatenation cost provides an estimate of an acoustic mismatch between units in the acoustic unit sequential pair.