Digital music synthesizers as perceptually valid phenomenological models

Abstract

Commercially available digital music synthesizers can produce sounds which perceptually resemble many traditional acoustic instruments. As with all phenomenological modeling, the quality of the output (of a synthesizer, in this case) does not require the generation process to incorporate the physics of the modeled systems (which are acoustic musical instruments in this case). Spectrum‐versus‐time and spectrum‐versus‐pitch plots for several commercially manufactured digital synthesizers employing different synthesis techniques (including audio‐rate frequency modulation) will be compared with their counterparts from traditional acoustic instruments. It is also of interest to compare the spectra after imposing transformations which model psychoacoustic processing, such as loudness curves and integration within critical bands.