An experimental SDIF-sampler in Max/MSP

Samplers are traditionally distinguished from synthesizers according to their input data: raw audio for samplers or model-based representation for synthesizers. We would like to introduce here a somewhat intermediary concept by taking advantage of the remarkable flexibility and power of a combined use of a real-time environment such as Max/MSP and an audio-signal model description standard such as SDIF (Sound Description Interchange Format). In short, our SDIF-sampler uses SDIF-files in a similar manner as a sampler uses audio-files. We will review the different stages implied in the construction of such a sampler namely analysis, re/synthesis and user-interface design. This concept though still in an experimental phase has been tested with a realistic set-up.