Waveset transformations: source characteristics and transformation peculiarities producing hardly predictable sound results

• Autor(a) principal: Fellipe Miranda Martins
• Data de Publicação: 2021
• Outros Autores: José Henrique Padovani Velloso
• Tipo de documento: Artigo de conferência
• Idioma: eng
• Título da fonte: Repositório Institucional da UFMG
• Texto Completo: https://doi.org/10.5281/zenodo.5113511; http://hdl.handle.net/1843/54833; https://orcid.org/0000-0002-2686-6203; https://orcid.org/0000-0002-8919-7393
• topic: Wavesets transformations; Waveforms; Sonorities; Digital audio effects; Enhancements; Ondas sonoras; Som - Transmissão; Som - Técnicas digitais

The wavesets are defined as a signal portion between three consecutive zero-crossings and were proposed by Trevor Wishart as a form to manipulate and transform sound files. For simple sounds, wavesets tend to coincide with the wave cycle, but for complex and polyphonic sounds the wavesets can represent a small portion of longer oscillations or retain a slow oscillation superposed by faster ones. As wavesets’ shape, duration and amplitude are strongly unique for each sound, modifying a waveform based on this granular criterion can lead to unique sonorities. Firstly, we discuss the main issues that cause highly different sound results when applying the same waveset transformation to different waveforms (e.g., DC, low-frequency content, beats, phase, and others), secondly, we selected several waveset transformations to evaluate the range of sonorities produced as well as contextualizing them in the field of wellknown digital audio effects. In the end, we evaluate the major pitfalls of working with this technique and report some enhancements that are currently being studied.