Title :
Linear Regression for Prosody Prediction via Convex Optimization
Author :
Cen, Ling ; Dong, Minghui ; Chan, Paul
Author_Institution :
Inst. for Infocomm Res. (I2R), A *STAR, Singapore, Singapore
Abstract :
In this paper, a L1 regularized linear regression based method is proposed to model the relationship between the linguistic features and prosodic parameters in Text-to-Speech (TTS) synthesis. By formulating prosodic prediction as a convex problem, it can be solved using very efficient numerical method. The performance can be similar to that of the Classification and Regression Tree (CART), a widely used approach for prosodic prediction. However, the computational load can be as low as 76% of that required by CART.
Keywords :
convex programming; numerical analysis; regression analysis; speech synthesis; L1 regularized linear regression; classification-and-regression tree; convex optimization; linguistic feature; numerical method; prosody prediction; text-to-speech synthesis; Convex functions; Hidden Markov models; Linear regression; Pragmatics; Speech; Speech synthesis; Vectors; Speech synthesis; convex optimization; linear regression; prosody prediction;
Conference_Titel :
Asian Language Processing (IALP), 2011 International Conference on
Conference_Location :
Penang
Print_ISBN :
978-1-4577-1733-8
DOI :
10.1109/IALP.2011.75
