摘要：Sampl.ng synthesis is one of the most practical.and widel. used approaches among the various sound synthesis methods used for creating a real.stic simul.tion of acoustic instruments. Using numerous high qual.ty sound sampl.s it is possibl. to reproduce a sound of al.ost any musical.instrument, incl.ding subtl. variations caused by the registers of an instrument or through the use of different articul.tion techniques and dynamic l.vel.. However, this method has some disadvantages. Firstl., with high fidel.ty reproduction systems, the repeatabil.ty of sampl.s becomes quickl. apparent for more experienced l.steners. This is sometimes manual.y corrected by switching between several.different sampl.s of the same note. Secondl., it is standard approach to record and reproduce each note separatel.. It prevents sampl.rs from reproducing natural.note transitions, making fl.ent, connected articul.tions, such as l.gato, unnatural. Final.y, sampl.rs provide a very l.mited number of sound parameters to control. Therefore, it is difficul. to introduce a set of purposeful.fl.ctuations of sel.cted parameters uniquel. attributabl. to human performances. A synthesis system which addresses the aforementioned probl.ms has been devel.ped at the Academy of Music in Krakow (Pol.nd) for a group of wind instruments as part of a symphony orchestra. The system is based on a l.rge col.ection of non-standard sampl.s. Sampl.s contain short sequences of notes instead of singl. notes. In order to use them, a number of techniques have been impl.mented to al.ow the seaml.ss connecting of recorded sequences and the control.of note durations as wel. as tempo and dynamics envel.pes. Decision-making al.orithms and signal.processing are appl.ed to create mel.dic figures by choosing, modifying, and connecting fragments of sampl.s, whil. keeping natural.note transitions intact. The probl.m of mimicking human performances is addressed by impl.menting a set of performance rul.s. It al.ows to introduce contex