Duration prediction modeling in speech synthesis

Details Duration prediction modeling in speech synthesis Duration prediction modeling in speech synthesis

2006-10-19

2010-11-23

Lerner, Martin (Department: 2626)

Data processing: speech signal processing, linguistics, language

Speech signal processing

Synthesis

C704S266000, C704S267000

Reexamination Certificate

active

07840408

ABSTRACT:
The present invention provides a method and apparatus for training a duration prediction model, method and apparatus for duration prediction, method and apparatus for speech synthesis. Said method for training a duration prediction model, comprising: generating an initial duration prediction model with a plurality of attributes related to duration prediction and at least part of possible attribute combinations of said plurality of attributes, in which each of said plurality of attributes and said attribute combinations is included as an item; calculating importance of each said item in said duration prediction model; deleting the item having the lowest importance calculated; re-generating a duration prediction model with the remaining items; determining whether said re-generated duration prediction model is an optimal model; and repeating said step of calculating importance and the following steps, if said duration prediction model is determined as not optimal model.

REFERENCES:
patent: 5561421 (1996-10-01), Smith et al.
patent: 5682501 (1997-10-01), Sharman
patent: 6038533 (2000-03-01), Buchsbaum et al.
patent: 6064960 (2000-05-01), Bellegarda et al.
patent: 6778960 (2004-08-01), Fukada
patent: 6810378 (2004-10-01), Kochanski et al.
patent: 6813604 (2004-11-01), Shih et al.
patent: 6934650 (2005-08-01), Yoshida et al.
patent: 7089186 (2006-08-01), Fukada
patent: 7412377 (2008-08-01), Monkowski
patent: 7457748 (2008-11-01), Nefti et al.
patent: 7643990 (2010-01-01), Bellegarda
patent: 2002/0165681 (2002-11-01), Yoshida et al.
patent: 2004/0083102 (2004-04-01), Nefti et al.
patent: 2004/0088723 (2004-05-01), Ma et al.
patent: 2005/0182630 (2005-08-01), Miro et al.
patent: 2006/0229877 (2006-10-01), Tian et al.
patent: 2007/0239439 (2007-10-01), Yi et al.
patent: 2007/0239451 (2007-10-01), Luan et al.
patent: 2007/0276666 (2007-11-01), Rosec et al.
patent: 2008/0059163 (2008-03-01), Ding et al.
patent: 2008/0082331 (2008-04-01), Luan et al.
patent: 2009/0171660 (2009-07-01), Jian et al.
“An RNN-Based Prosodic Information Synthesizer for Mandarin Text-to-Speech,” Sin-Horng Chen, et al., IEEE Transactions on Speech and Audio Processing, vol. 6, No. 3, May 1998, pp. 226-239.
“Polynomial Regression Model for Duration Prediction in Mandarin,” Sun Lu, et al., Iflytek Speech Laboratory, University of Science and Technology of China, 4 pages, Interspeech 2004, pp. 769 to 777.
“Linguistic Factors Affecting Timing in Korean With Application to Speech Synthesis,” Hyunsong Chung, et al., Department of Phonetics and Linguistics, University College London, U.K., 4 pages, Eurospeech 2001,815-819.
“Modeling Vowel Duration for Japanese Text-To-Speech Synthesis,” Jennifer J. Venditti, et al, Bell Labs—Lucent Technologies, Ohio State University, 4 pages, ICSLP 1998, pp. 786-789.

Affiliated with

Also associated with

Rating

Duration prediction modeling in speech synthesis does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Duration prediction modeling in speech synthesis, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Duration prediction modeling in speech synthesis will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-4155088