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. |
Year | DOI | Venue |
---|---|---|
2010 | 10.1109/ISCSLP.2010.5684871 | ISCSLP |
Keywords | Field | DocType |
natural speech synthesis,automatic prosodic break detection,support vector machine (svm),mandarin prosodic break detection,neural network (nn),syntactic evidence,conditional random fields (crfs),boosting classification and regression tree (cart),speech synthesis,complementary model,speech understanding,natural language processing,lexical evidence,acoustic evidence,acoustics,support vector machine,mathematical model,conditional random field,artificial neural networks,boosting,neural network | Computer science,Artificial intelligence,Natural language processing,Baseline system,Artificial neural network,Syntax,Speech synthesis,Pattern recognition,Neural network nn,Speech recognition,Classification tree analysis,Boosting (machine learning),Mandarin Chinese | Conference |
Volume | Issue | ISBN |
null | null | 978-1-4244-6244-5 |
Citations | PageRank | References |
0 | 0.34 | 5 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Chong-Jia Ni | 1 | 20 | 4.84 |
Wenju Liu | 2 | 214 | 39.32 |
Bo Xu | 3 | 1 | 1.03 |