Abstract | ||
---|---|---|
In this study we propose a data-based approach to intonationmodeling using vector quantization. The model is based onan F0 parametrization with an especially designed approximationfunction. The parameter vectors found are vectorquantized with varying codebook sizes. This method is motivatedby intonation theories that suggest that pitch accentand boundary phenomena can be described by a distinctnumber of different types. We use classification trees to predictthe F0 movements represented... |
Year | Venue | Keywords |
---|---|---|
1998 | SSW | classification tree,parametric model |
Field | DocType | Citations |
Parametric model,Parametrization,Pattern recognition,Computer science,Learning vector quantization,Pitch accent,Vector quantization,Quantization (physics),Artificial intelligence,Approximation function,Codebook | Conference | 19 |
PageRank | References | Authors |
1.72 | 5 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Alistair Conkie | 1 | 264 | 38.03 |
Gregor Mohler | 2 | 19 | 1.72 |