Synthesizers are divided by number of voices that are performed simultaneously and the way of signal processing. Monophony (performance one voice at the time) is used for recording in a studio. Poliphony (performance two or more voices at the time) is used for live performance. According to the method of signal procesing synthesizer are divided on analog (beginning of signal processing) and digital (current signal processing). Other types of synthesizers are: modulars (the first devices for generating audio effects, oversized and expensive), fully variables (the first commercial devices), presets (memory appearance, factory recorded sounds), programmables (many types of memory, possibility of deleting and recording sounds), samplers (the best type of devices, recording original sounds, parameter change of input signal for generating different effects), dedicated computers (central processing unit, the appearance of monitor for visual display, large speed of data processing), sound cards (modern computers, generating audio effects in real time, video games). Because of the various types of synthesis that synthesizer are used, every device have had there own way of work. Types of synthesis that can be used are: additive, subtractive, frequency modulation, phase distortion, linear-arithmetic, vectors, physical modulation, granular, wavetable. At the ned of the thesis there is suggestion of solution for the realization of the system for generating audio effects on a digital computer. Concepts for relization on digital computer written above are shown with different block schema and their description. There are also presented several examples of system operation on input signal processing and result of output signal. Finally, it is given review of advantages and disadvantages of our system.