SYSTEM AND METHOD FOR GENERALIZED PRESELECTION FOR UNIT SELECTION SYNTHESIS

• US 20110071836A1
• Filed: 09/21/2009
• Published: 03/24/2011
• Est. Priority Date: 09/21/2009
• Status: Active Grant

• Current Assignee: Cerence Operating Company (Cerence Inc.)
• Original Assignee: AT&T Intellectual Property I LP (AT&T, Inc.)
• Inventors: BEUTNAGEL, Mark, CONKIE, Alistair D., SYRDAL, Ann K., KIM, Yeon-Jun
• Granted Patent: US 8,805,687 B2
• Time in Patent Office: Days
• US Class Current: 704/267
• CPC Class Codes: G10L 13/00 Speech synthesis; Text to speech systems
• CPC Class Codes: G10L 13/047 Architecture of speech synthesisers
• CPC Class Codes: G10L 13/06 Elementary speech units used in speech synthesisers; Concatenation rules

Abstract

Disclosed herein are systems, computer-implemented methods, and computer-readable storage media for unit selection synthesis. The method causes a computing device to add a supplemental phoneset to a speech synthesizer front end having an existing phoneset, modify a unit preselection process based on the supplemental phoneset, preselect units from the supplemental phoneset and the existing phoneset based on the modified unit preselection process, and generate speech based on the preselected units. The supplemental phoneset can be a variation of the existing phoneset, can include a word boundary feature, can include a cluster feature where initial consonant clusters and some word boundaries are marked with diacritics, can include a function word feature which marks units as originating from a function word or a content word, and/or can include a pre-vocalic or post-vocalic feature. The speech synthesizer front end can incorporates the supplemental phoneset as an extra feature.

Claims

1. A method of generalized preselection for unit selection synthesis, the method causing a computing device to perform steps comprising:

adding a supplemental phoneset to a speech synthesizer front end having an existing phoneset;

modifying a unit preselection process based on the supplemental phoneset;

preselecting units from the supplemental phoneset and the existing phoneset based on the modified unit preselection process; and

generating speech based on the preselected units.

2. The method of claim 1, wherein the supplemental phoneset is a variation of the existing phoneset.

3. The method of claim 1, wherein the supplemental phoneset includes a word boundary feature.

4. The method of claim 1, wherein the supplemental phoneset includes a cluster feature where initial consonant clusters and some word boundaries are marked with diacritics.

5. The method of claim 1, wherein the supplemental phoneset includes a function word feature which marks units as originating from a function word or a content word.

6. The method of claim 1, wherein the supplemental phoneset includes a pre-vocalic or post-vocalic feature.

7. The method of claim 1, wherein the speech synthesizer front end incorporates the supplemental phoneset as an extra feature.

8. The method of claim 7, wherein preselecting units further comprises assigning costs to units in one phoneset based on whether a unit of interest agrees in terms of another phoneset.

9. A system for generalized preselection for unit selection synthesis, the system comprising:

a processor;

an adding module configured to control the processor to add a supplemental phoneset to a speech synthesizer front end having an existing phoneset;

a modification module configured to control the processor to modify a unit preselection process based on the supplemental phoneset;

a preselection module configured to control the processor to preselect units from the supplemental phoneset and the existing phoneset based on the modified unit preselection process; and

a generation module configured to control the processor to generate speech based on the preselected units.

10. The system of claim 9, wherein the supplemental phoneset is a variation of the existing phoneset.

11. The system of claim 9, wherein the supplemental phoneset includes a word boundary feature.

12. The system of claim 9, wherein the supplemental phoneset includes a cluster feature where initial consonant clusters and some word boundaries are marked with diacritics.

13. The system of claim 9, wherein the supplemental phoneset includes a function word feature which marks units as originating from a function word or a content word.

14. The system of claim 9, wherein the supplemental phoneset includes a pre-vocalic or post-vocalic feature.

15. The system of claim 9, wherein the speech synthesizer front end incorporates the supplemental phoneset as an extra feature.

16. The system of claim 15, wherein the preselection module is further configured to control the processor to assign costs to units in one phoneset based on whether a unit of interest agrees in terms of another phoneset.

17. A computer-readable storage medium storing instructions which, when executed by a computing device, control the computing device to perform generalized preselection for unit selection synthesis, the instructions comprising:

adding a supplemental phoneset to a speech synthesizer front end having an existing phoneset;

modifying a unit preselection process based on the supplemental phoneset;

preselecting units from the supplemental phoneset and the existing phoneset based on the modified unit preselection process; and

generating speech based on the preselected units.

18. The computer-readable storage medium of claim 17, wherein the supplemental phoneset is a variation of the existing phoneset.

19. The computer-readable storage medium of claim 17, wherein the supplemental phoneset includes a word boundary feature.

20. The computer-readable storage medium of claim 17, wherein the supplemental phoneset includes a cluster feature where initial consonant clusters and some word boundaries are marked with diacritics.