Project

Sound Design with Everyday Words

Sound designers, audio professionals, and musicians often spend time and energy looking for the right sound for a particular piece of music or sonic environment. Current sound synthesizers either contain numerous sound presets that are laborious to parse, or batteries of parameters to tweak without straightforward connections to one's intuitive expectation. We propose a sound retrieval and modification engine based on everyday words like "bright," "warm," and "fat." The perceptual sound synthesis engine is informed by a survey of musicians and listeners worldwide and can also be customized. This system allows dynamic tagging of sound material from online libraries, and "sound sculpting" based on common verbal descriptors instead of obscure numerical parameters.

Sound designers, audio professionals, and musicians often spend time and energy looking for the right sound for a particular piece of music or sonic environment. Current sound synthesizers either contain numerous sound presets that are laborious to parse, or batteries of parameters to tweak without straightforward connections to one's intuitive expectation. We propose a sound retrieval and modification engine based on everyday words like "bright," "warm," and "fat." The perceptual sound synthesis engine is informed by a survey of musicians and listeners worldwide and can also be customized. This system allows dynamic tagging of sound material from online libraries, and "sound sculpting" based on common verbal descriptors instead of obscure numerical parameters.