Title | ||
---|---|---|
CART-based modeling of Chinese tonal patterns with a functional model tracing the fundamental frequency trajectories |
Abstract | ||
---|---|---|
We propose an approach to modeling Chinese tonal patterns, focusing on the basic fundamental frequency (F0) patterns characterized by the contextual linguistic features that can be directly extracted from text. We analyze tonal patterns as sparse target points (tonal F0 peaks and valleys) and represent them in parametric form within the framework of a functional F0 model. The relationships between the target points and underlying linguistic features are trained using classification and regression tree analysis (CARTs), and this functional model is used to trace the F0 trajectories when training the CARTs and to synthesize a tonal pattern from the target points predicted by the CARTs. Our experiments indicate that the proposed method has low F0 prediction errors. Utilization of the F0 ranges measured from training samples could significantly reduce the influences of differences in voice ranges on training a speaker-independent model. Furthermore, the most important roles in characterizing tonal patterns were played by a few linguistic features such as lexical tone context and the distinction between voiced from unvoiced initials. |
Year | DOI | Venue |
---|---|---|
2009 | 10.1109/ICASSP.2009.4960568 | ICASSP |
Keywords | Field | DocType |
contextual linguistic feature,chinese tonal pattern,f0 prediction error,f0 range,functional model,fundamental frequency trajectory,cart-based modeling,f0 peak,tonal pattern,f0 model,target point,f0 trajectory,context modeling,data mining,natural languages,speech processing,prediction error,frequency,pattern analysis,speech,feature extraction,fundamental frequency,hidden markov models,predictive models,correlation,learning artificial intelligence,machine learning,speech synthesis | Speech processing,Speech synthesis,Parametric equation,Fundamental frequency,Pattern recognition,Computer science,Feature extraction,Speech recognition,Correlation,Artificial intelligence,Hidden Markov model,Tracing | Conference |
ISSN | Citations | PageRank |
1520-6149 | 0 | 0.34 |
References | Authors | |
3 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Jinfu Ni | 1 | 88 | 16.32 |
Shinsuke Sakai | 2 | 126 | 23.52 |
Tohru Shimizu | 3 | 57 | 12.85 |
Satoshi Nakamura | 4 | 1099 | 194.59 |