Algorithms created by me

PadSynth synthesis algorithm

The PadSynth algorithm is designed to produce beautiful sounds and, unlike other synthesis algorithms, it produces more "natural" sounds. It works best on creating ensemble sounds like choirs or string pads.

PADsynth audio demo

PADsynth steps:

A more detailed description of the algorithm is available and it was published in the paper: Paul Nasca, “Padsynth” Sound Synthesis Algorithm, STUDIA UNIV. BABES–BOLYAI, INFORMATICA, Volume LVI, Number 1, 2011.

This algorithm is implemented in many commercial and open source software synthesizers:

"PaulStretch" extreme sound stretching algorithm

This algorithm is designed to stretch the sound by a very large amount while preserving the subjective quality. For example using this algorithm it is possible to stretch 5 minutes of sound to 1 hour. The best results are on orchestral/choir audio inputs.

PaulStretch steps:

This algorithm is implemented in Paul's Extreme Sound Stretch (in C++ and Python languages) and it is included as an effect in Audacity.


This algorithm converts the audio to images and vice-versa. The sound or image is processed in a single block. It is implemented in Hypermammut software.

LDR Tonemapping

LDR tonemapping algorithm was inspired by an effect for GIMP. It enhance the details of the image and increases the local contrast. It was implemented in my own software with the same name and as an efect for digiKam.


This is a weird variant of the "wah-wah" audio effect. It is implemented in ZynAddSubFX and as the LADSPA effect.