Blanc SIMI 2 - Blanc - SIMI 2 - Science informatique et applications

Faust Environment Everyware: an ubiquitous solution for multiplatform and portable numerical audio processing – FEEVER

Submission summary

Music has been transcending societies and eras. In our digital world, from portable MP3 players to radio streaming to Surround Sound-equipped movie theaters, music and more generally audio processing is a life-enhancing practice that impacts everyone everywhere in evermore personalized manners. Yet, the Net has been slow to offer users and developers the kind of advanced technologies that would make similar listening enrichment standard, seamless and easily customizable online. If HTML-based Web sites can easily be equipped with advanced general processing capabilities via scripting languages such as Javascript or graphically enhanced via powerful display tools such as WebGL, no such technology exists in-the-large in the audio domain: the FEEVER project aims at providing the key technologies that will make the whole digital world sing, be it when web surfing at home or listening to a radio stream on the go.
Imagine a world in which an audio engineer would design a new reverberation algorithm and post its implementation somehow on the Internet. A listener, looking for a richer audio experience, could use it by browsing the corresponding site, which would automatically lead to the download and activation of the corresponding sound processing snippet in her browser or even reroute her audio stream to the remote reverb site for processing. Equipping the reverb site with remote management capabilities paves the way to even richer sound installations in which multiple processing platforms would collaborate to provide to the user a dedicated listening environment; a couple of touches on her tablet, and audio parameters could be tuned to suit a changing situation, for instance when moving from the comfort of her home to the noisier environment of her car.
FEEVER intends to make such a vision a reality. Yet, scientific and technical challenges abound: the technological solutions need to be (1) portable, to allow program-once, deploy-everywhere economic advantages, (2) easily programmable, to narrow the gap between specifications and implementation, (3) able to deal with multiple platforms, for seamless integration within the users listening environments, (4) efficient both in terms of computing time, since audio processing is a highly compute-intensive activity, and energy, if only to permit mobile applications, and (5) secure, since audio processing activity performed on the client side must not jeopardize the user system.
We believe the Faust programming ecosystem introduced at GRAME 10 years ago is the proper starting point towards the global solution the audio world is waiting for. FEEVER intends to make this happen via five tasks, Task 0 being dedicated to project management. Task 1 focuses on formal issues at the language level: Faust, built upon the functional and synchronous programming paradigms, is a mono-rate-only specification paradigm, which makes processing algorithms such as FFT cumbersome. We intend to offer a multi-rate solution, a key design that will also help when dealing with user interactions, which occur at rates different from audio. Task 2 intends to provide an industrial-strength, efficient, multi-rate, multi-platform, portable, easily integrable Faust compiler; looking at validation issues here is key in order to ensure that the provided solution is secure. Task 3 puts FEEVER on the global scene, looking at ways to make Faust-enabled solutions available everywhere, be it as a Web service or integrated within a smartphone web browser. Finally, Task 4 strives to make FEEVER technologies even more relevant by looking at the new usage they open up for audio processing and, more generally, digital signal processing teaching; Faust, already taught in some world renown institutions, and FEEVER-derived technologies should lead to new, more interactive and fun ways to teach these key engineering and artistic subjects, further fostering interdisciplinary collaborations between the art and scientific communities.

Project coordination

Pierre Jouvelot (ARMINES Centre de Recherche en Informatique de Mines ParisTech)

The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.

Partner

GRAME, Centre National de Création Musicale
Inria Rennes - Bretagne Atlantique Inria, Centre de recherche de Rennes - Bretagne Atlantique
UJM/CIEREC CIEREC
ARMINES CRI ARMINES Centre de Recherche en Informatique de Mines ParisTech

Help of the ANR 499,743 euros
Beginning and duration of the scientific project: November 2013 - 42 Months

Useful links

Explorez notre base de projets financés

 

 

ANR makes available its datasets on funded projects, click here to find more.

Sign up for the latest news:
Subscribe to our newsletter