| 摘要 |
An improved control structure for music synthesis is provided in which: 1) the sound generation provided to the adaptive function mapper (501) allows for a greatly increased degree of control over the sound produced; and 2) training of the adaptive function mapper (501) is performed using an error measure, or error norm, that greatly facilitates learning while ensuring perceptual identity of the produced sound with the training example. In one embodiment, sound data is produced by applying to an adaptive function mapper (501) control parameters including: at least one parameter selected from the set of time and timbre space coordinates; and one parameter from the set of pitch, delta pitch, articulation and dynamic. Mapping is performed from the control parameters to synthesis parameters to be applied to a sound synthesizer. In another embodiment, an adaptive function mapper (501) is trained to produce, in accordance with the information stored in a mapping store (503), synthesis parameters to be applied to a sound synthesizer, by steps including: analyzing sounds to produce sound parameters describing the sounds; further analyzing the sound parameters to produce control parameters; applying the control parameters to the adaptive function mapper (501), the adaptive function mapper (501) in response producing trial synthesis parameters comparable to the sound parameters; deriving from the sound parameters and the trail synthesis parameters an error measure in accordance with a perceptual error norm in which they are weighted in approximate degree to which they are perceived by the human ear; and adapting the information in the mapping store (503) with the error measure.
|
