Title :
Mandarin prosodic break detection based on complementary model
Author :
Ni, Chong-Jia ; Liu, Wen-Ju ; Xu, Bo
fDate :
Nov. 29 2010-Dec. 3 2010
Abstract :
Automatic prosodic break detection is important for both speech understanding and natural speech synthesis. In this paper, we develop complementary model to detect Mandarin prosodic break by using acoustic, lexical and syntactic evidence. The model realizes the complementarities by taking the advantages of each model. When comparing with the baseline system, our proposed method has good performance.
Keywords :
natural language processing; speech synthesis; Mandarin prosodic break detection; acoustic evidence; automatic prosodic break detection; complementary model; lexical evidence; natural speech synthesis; speech understanding; syntactic evidence; Acoustics; Artificial neural networks; Boosting; Classification tree analysis; Equations; Mathematical model; Syntactics; boosting classification and regression tree (CART); complementary model; conditional random fields (CRFs); neural network (NN); support vector machine (SVM);
Conference_Titel :
Chinese Spoken Language Processing (ISCSLP), 2010 7th International Symposium on
Conference_Location :
Tainan
Print_ISBN :
978-1-4244-6244-5
DOI :
10.1109/ISCSLP.2010.5684871
