Design and Run Real-time Spectral Processing on the Web with Faust

Shihong Ren (1) , Stéphane Letz (2) , Michel Buffa (3, 4, 5) , Laurent Pottier (1) , Yang Yu (6)

1 UJM - Université Jean Monnet - Saint-Étienne
2 GRAME - Centre national de création musicale [Lyon]
3 WIMMICS - Web-Instrumented Man-Machine Interactions, Communities and Semantics
4 I3S - Laboratoire d'Informatique, Signaux, et Systèmes de Sophia Antipolis
5 UniCA - Université Côte d'Azur
6 shangaï conservatory of music

Web-based spectral processing with the Web Audio API is a challenging task that requires efficient and flexible tools. It involves Fourier transform utilities and frequency-domain data manipulations. In this paper, we present a novel frame- work for designing and running real-time spectral processors on the web, using Faust as the programming language and its web-based toolchain. Our framework is inspired by Max’s pfft~ paradigm, which allows users to create custom spec- tral processors from streamed spectral data. Using Faust language and suitable FFT tools, the designed algorithm can be compiled to WebAssembly modules that are executed in an Audio Worklet. We demonstrate the potential of our framework by showing some examples of spectral effects and synthesizers that can be easily designed and executed on the web.

Domaines: Web; Multimédia [cs.MM]

Shihong Ren, Stéphane Letz, Michel Buffa, Laurent Pottier, Yang Yu. Design and Run Real-time Spectral Processing on the Web with Faust. WAC 2024 - Web Audio Conference 2024, Purdue University / Tae Hong Park, Mar 2024, Lafayette, Indiana, United States. ⟨10.5281/zenodo.10825715⟩. ⟨hal-04507625⟩