Analysis of Concatenative Synthesis Techniques

Concatenative synthesis is a technique of Sound Texture Synthesis where a short clip of source audio is used to generate a continuous stream of similar sound. Several approaches for concatenative synthesis exist, but their output quality is highly dependent on the type and length of the source material, as well as the parameters used. The purpose of this paper is to test different synthesis techniques, parameters, and types of source material, to determine which algorithms best suit a variety of input sources. The study reveals that in most cases the complexity of the algorithm has a relatively small effect on the output quality. Noisy and repetitive sources produce the most natural sounding outputs even with short input clips and atom sizes; whereas, sources with a lot of variation over time typically require larger atom sizes and more source material to prevent audible artefacts in the synthesis output.