AUSSCHREIBUNG - Researchers Position, Ircam, Quaero Project

Von: DEGEM / Cathy van Eck
From: Aki Pasoulas

From April 1st, 2010

Introduction to IRCAM

IRCAM is a leading non-profit organization associated to Centre  
Pompidou, dedicated to music production, R&D and education in  
acoustics and music. It hosts composers, researchers and students from  
many countries cooperating in contemporary music production,  
scientific and applied research. The main topics addressed in its R&D  
department include acoustics, audio signal processing, computer music,  
interaction technologies, musicology. Ircam is located in the centre  
of Paris near the Centre Pompidou, at 1, Place Igor Stravinsky 75004  
Paris.

Introduction to Quaero project

Quaero is a 200 M€ collaborative research and development program  
focusing on the areas of automatic extraction of information,  
analysis, classification and usage of digital multimedia content for  
professionals and consumers. The research work shall concentrate on  
managing virtually unlimited quantities of multimedia and multilingual  
information, including text, speech, music, image and video. Five main  
application areas have been identified by the partners:

1.       multimedia internet search

2.       enhanced access services to audiovisual content on portals

3.       personalized video selection and distribution

4.       professional audiovisual asset management

5.       digitalization and enrichment of library content, audiovisual  
cultural heritage and scientific information.

The Quaero consortium was created to meet new multimedia content  
analysis requirements for consumers and professionals, faced with the  
explosion of accessible digital information and the proliferation of  
access means (PC,  TV, handheld devices). More information can be  
found at www.quaero.org/.

Role of Ircam in Quaero Project

In the Quaero project, Ircam is in charge of the coordination of audio/
music indexing research and of development of music-audio indexing  
technology: music content-description (tempo, rhythm, key, chord,  
singing-voice, and instrumentation description), automatic indexing  
(music genre/style, mood), music similarity, music audio summary,  
chorus detection and audio identification. A specificity of the  
project is the creation of a large-music-audio corpus in order to  
train and validate all the algorithms developed during the project.

Position description

Researchers would be in charge of the development of the technologies  
related to

·         music-audio content description/ content-extraction: tempo  
and beat/measure position estimation, key/mode, chord progression,  
instrument/drum identification, singing voice location, voice  
description

·         music automatic indexing into music genre, music mood

·         music similarity: especially on large-scale databases

·         music structure discovery, automatic music audio summary  
generation, chorus location

Researcher will also collaborate with the evaluation team who evaluate  
algorithms performances and with the developer team.

Required profile

·         Very high skills in audio signal processing (spectral  
analysis, audio-feature extraction, parameter estimation)

·         High skill in audio indexing and data mining (statistical  
modelling, automatic feature selection algorithm, …)

·         High-skill in large-database search algorithms

·         Good knowledge of Linux, Windows, MAC-OS environments

·         High-skill in Matlab programming, skills in C/C++ programming

·         High productivity, methodical work, excellent programming  
style.


Salary

According to background and experience

Applications

Please send an application letter together with your resume and any  
suitable information addressing the above issues preferably by email  
to: peeters_a_t_ircam dot fr with cc to vinet_a_t_ircam dot fr,  
rod_a_t_ircam_dot_fr, roebel_at_ircam_dot_fr