In the context of the MOSART Program, a software-based realtime saxophone synthesizer is being developed at the Music Technology Group of the IUA. Main objectives are high sound quality and flexible control over meaningful musical performance parameters. The synthesis concept is built upon the “Sinusoidal plus Residual Model”. A database holds spectrally pre-analyzed sound segments, which are used as anchor-points in a continuous multi dimensional timbre space. Each spectral segment consists of a stationary sinusoidal and a non-stationary residual part. By applying spectral interpolation techniques and modeling specific spectral properties, a realistic timbre space can be synthesized. With this method it is possible to preserve the “sample-like” quality when needed by simply adding both parts together without applying any modification to them. By treating the parts individually the whole range and flexibility of spectral transformation and interpolation techniques becomes available for the synthesis process.
[1]
Xavier Serra,et al.
Integrating complementary spectral models in the design of a musical synthesizer
,
1997,
ICMC.
[2]
Marcelo M. Wanderley,et al.
Instrumental Gestural Mapping Strategies as Expressivity Determinants in Computer Music Performance
,
1997
.
[3]
Larry Polansky,et al.
Spectral Mutation in Soundhack
,
1996
.
[4]
Xavier Serra,et al.
A system for sound analysis/transformation/synthesis based on a deterministic plus stochastic decomposition
,
1989
.
[5]
Malcolm Slaney,et al.
Automatic audio morphing
,
1996,
1996 IEEE International Conference on Acoustics, Speech, and Signal Processing Conference Proceedings.
[6]
Xavier Rodet,et al.
New Applications of the Sound Description Interchange Format
,
1998,
ICMC.
[7]
Xavier Serra,et al.
Musical Sound Modeling with Sinusoids plus Noise
,
1997
.
[8]
Roger B. Dannenberg,et al.
Synthesizing Trumpet Performances
,
1998,
ICMC.