EEG.pdf


Aperçu du fichier PDF eeg.pdf - page 1/31

Page 12331



Aperçu texte


ELECTRONIC AND EMOTIVE SOUND PROCESSING
GODEFROID Robin
1st Master Electronic
Robin.Godefroid@Gmail.com
A BSTRACT

2. SYNOPSIS

The project, shown in this document, offers a new perspective for sound processing, in the arts, based on use of a
sensory sensor and a rapid prototyping software.
An electroencephalography headset [9] is used in order to recognize user’s feelings. These information are
retrieved through the open-source workbench Skemmi
[10]. Furthermore, this workbench includes the components needed to communicate with Kyma [11], a powerful
sound processing software.
The work was to implement, inside Skemmi, tools that
allows a simple control, requiring a minimum technical
prerequisites for users, of the sounds settings like the volume, the distorsion, etc.
The application was developed in an open way to allow
use of other sensory inputs interface like, for example, a
Wiimote or a Kinect.
Keywords Signal processing · Sound processing · Emotiv
Epoc · Kyma · OpenInterface · human/machine interface ·
Electroencephalography · C++
1. INTRODUCTION
Nowadays, creating music is mostly based on two approaches :
• A “classical” approach in which music comes from
“ancestral” instruments and is played, directly, by
the artists ;
• An electronic approach where the composition is
done through a computer or a synthesizer.
However, with technological advances which allow
more and more efficient and affordable solutions for sensory sensors (motion detection, sound sensor, . . . ), and
modern computing tools that permit a simple use of these
interfaces, we can imagine a new way for conceiving music.
This one would allow a interaction mainly based on
user’s behavior, his gestures, ... which would give other
artistic possibilities in musical composition.
It’s with this idea that our work has started. We decided
to study the possibility of using an electroencephalographic headset to recognize the user/artist’s feelings in order
to allow sound and sound effects modification according
to these collected information.

The information, about the user’s behavior, are transmited, via a wireless connection, from the electroencephalographic headset to Skemmi. The communication between Skemmin (running on a computer) and Kyma (running on a MAC) is established through Internet using the
OSC protocol 1 . Thereby forwarded to Kyma, the desired
action is finally “materialized” (≡ sonification) through
electronic processing done by the Pacarana, and the audio
interface.
3. SPECIFICATIONS
The application has to allow :
• an interaction as simple as possible with the sound
processing software ;
• an automatic update of the sound settings (dynamic
adaptation) ;
• interfacing the main electroencephalographic headset functions (depending on the API 2 available).

1 . OpenSoundControl, data transmission protocol through Internet
network
2 . ApplicationProgrammingInterface