AudioGarden: A Usable Tool for Composite Audio Creation

To content | To menu | To search

Monday 2 August 2010

Companion video for the AudioMostly 2010 conference

Tuesday 29 June 2010

First results will be presented at Audio Mostly 2010!

logo-audiomostlyBackground.png

Our first results concerning our project AudioGarden will be presented in a poster entitled Towards User-friendly Audio Creation at the Audio Mostly 2010 Conference, next September in Pitea, Sweden.

The Audio Mostly Conference provide a venue to explore the unexploited potential of audio in computer-based environments, for example game contexts, and aim to help open up this area of thinking by bringing together game designers, audio experts, content creators, and technology and behavioral researchers.

Our four page article will be published in the ACM Digital Library.

Saturday 22 May 2010

Synthesis method

The method consists in automatically segmenting available audio recordings into atomic grains, i.e., small signals from 0.01 to 0.1 sec. Audio grains are then classified by similarity in our navigation tool, an extended version of AudioCycle. By selecting one recording as a model for rhythmic pattern and timbre evolution and a set of audio grains, our synthesis procedure creates new sound events based on onset detection and similarity measurements. The figure below illustrates the technique: the original recording in red is taken as a model for the synthesized one (in blue); the grains identity are specified by their number, and dotted lines show their boundaries.

Synthesis.png

Sunday 2 May 2010

Preliminary mockups of the user interface

To support our workflow, we had to design a dedicated user interface. To achieve a certain level of mutual understanding of what we believe to be a suitable design, we produced throughout several brainstormings many mockups of the visual user interface. Examples are given below.

100401-ConcentricCrownViz_JogWheel-WhiteboardStoryboardLighter.jpg

AudioGarden_L1b.png

Wednesday 14 April 2010

Browsing and manipulating audio loops with gestural controllers

The AudioGarden prototype is in progress, based on top of the MediaCycle framework developed within the numediart Research Program in Digital Art Technologies in Belgium, towards hypermedia navigation. We are now investigating which are the best controller candidates to be hooked up to our AudioGarden prototype.

Tuesday 6 April 2010

First working week in Belgium

Last week (from March 29th to April 2nd), we organised a working week in Belgium, in order to set up "in vivo" the main requirements of the AudioGarden project. We mainly focused on the user interface by producing some sketches, mockups and storyboards. Here is one of the locations where we installed our "office"..

IMG_0618.JPG

Friday 2 April 2010

numediart call for participation

numediart is a long-term research program centered on Digital Media Arts, funded by Région Wallonne, Belgium (grant N°716631). Its main goal is to foster the development of new media technologies through digital performances and installations, in connection with local companies and artists.

cfp10_cover1.JPG

AudioGarden has been proposed as a numediart short project, to be held (remotely) from April, 5th to June, 30th 2010, and the call for participation has been published. Thus, Damien Tardieu joins the project as collaborator.

Thursday 1 April 2010

Presentation at TCTS Lab, University of Mons (Be)

On Monday started the visiting week to work with Christian on AudioGarden project. Today, I have been invited to present my work on sound modeling to the TCTS Laboratory in Mons (Be). Since some of the research here is dedicated to rendering of virtual environments and sound analysis/synthesis, the presentation raised many questions!

Friday 26 March 2010

AudioGarden at Telekom Labs - Design Research, Berlin

I have been invited by the Telekom Labs - Design Research in Berlin (Ge) to present my work on sound modeling on the 22nd of March. This was a good opportunity to introduce our current research on AudioGarden and its objectives. People are waiting for results! We hope to soon present a prototype of our tool.

Thursday 4 February 2010

(Opensource) 3D Modelers

We aim at gathering physically based synthesis methods and handling of prerecordings in the same interface. The idea will be to address physically based synthesis through modal analysis, which consists in modeling a vibrating object by a bank of damped harmonic oscillators excited by an external stimulus. Modal decomposition allows efficient runtime and control for vibrational response. Modal sounds can be considered themselves as audio grains, and they can be consistently introduced into the database classification. Thus, the sound material becomes composite, extending the rendering of the synthesis output.

Physically based sounds may allow the user to design from a shape, as for example with a CAD model, to its sound and incorporate it into a sound composition. For this purpose, we are looking for a tool that mainly focuses on shaping/sculpting geometries easily, quickly, efficiently. Alias|Wavefront *.obj format support is mandatory.

Continue reading...

Wednesday 3 February 2010

Our application to a COST-SID STSM grant has been approved!

Our Short-Term Scientific Missions (STSM) application to the Sonic Interaction Design (SID) COST Action has been accepted. It will allow us to fund Cecile's visit to the research labs partners of this project in Belgium, at UCL-TELE and UMons/TCTS, within the numediart research program on digital art technologies.

Wednesday 6 January 2010

New Year Resolution: AudioGarden!

Best wishes for the new year!

Our prime resolution for this year is to release a first working prototype of AudioGarden. Get more info on this project on the About page.

Stay tuned for upcoming news!