Philippe ESLING

PhD Student
IRCAM - CNRS UMR 9912
Supervisor : Carlos Agon
Music Representations
esling@ircam.fr

* Intelligent sound samples database : Seeking sound samples in a massive database can be a very tedious and painstaking task. Even when meta-informations are available, querying results may remain far from the mental representation expected by the user. We worked on the development of the first intelligent sound sample database. We propose a scheme where sounds can be retrieved by simply drawing the temporal shape of spectral descriptors. Then we addressed two completely novel ways of intuitive querying. First, by optimizing simultaneously multiple temporal shapes of descriptors with MultiObjective Spectral Evolution Query (MOSEQ). Second by performing a vocal imitation of the sound sample with Query by Vocal Imitation (QVI).

* Multiobjective time series matching : We present here an innovative problem that can be casted into a new approach merging multiobjective optimization and time series matching algorithms, called MultiObjective Time Series (MOTS) matching. We formally state this novel problem that could lead to a whole range of applications in several fields of research and report an efficient implementation. This approach allowed in the scope of sound samples querying to cope with the multidimensional nature of timbre perception and also to obtain a whole set of efficient propositions rather than a single best solution.

* Musical orchestration : Orchestration is the subtle art of mixing instrumental properties. Among all techniques of musical composition, it has always remained an empirical activity. Our approach is intended to search for sound combinations within instrument sample databases that best match a target timbre defined by the composer. We propose an original approach for the discovery of relevant sound combinations, in which we explicitly address combinatorial issues and tackle the problem of temporal descriptors evolution.

• ATO-MS : Abstract Temporal Orchestration - Modular Structure : The new beta release of the temporal orchestration software which allows complex interactions with time-evolving structure in order to produce rich orchestrations.

• MultiObjective Time Series (MOTS) matching for audio classification and retrieval from IEEE Transactions on Audio, Speech and Language Processing (TSALP)

• EDITE Grant 2009 for doctoral studies
• AFIM Prize 2010 of the young researcher
• Tokyo Wonder Site 2011 Experimental Music Festival

• Esling Philippe, Agon Carlos "Time series data mining and analysis", ACM Computing Surveys 2012 (Accepted proof).
• Lecroq Béatrice, Lejzerowicz Franck, Esling Philippe, Baerlocher Loic, Farinelli Laurent, Pawlowski Jan "Ultra-deep sequencing of foraminiferal microbarcodes unveils hidden richness of early monothalamous lineages in deep-sea sediments", Publication of the National Academy of Science, vol.108, no.32, pp 13177-13182, August 2011.
• Esling Philippe, Agon Carlos "Multiobjective time series matching for audio classification and retrieval", IEEE Transactions on Speech Audio and Language Processing 2012 (in review).

• Esling Philippe, Carpentier Grégoire, Agon Carlos "Dynamic Musical Orchestration using Genetic Algorithms and a Spectro-Temporal Description of Musical Instruments", Lecture Notes in Computer Science, vol. 6025, EvoApplications Part II, 2010.

• Esling Philippe, Agon Carlos "Composition of Sound Mixtures with Spectral Maquettes", Proceedings of the International Computer Music Conference, New York, USA, 2010.
• Esling Philippe, Carpentier Grégoire, Agon Carlos "Dynamic Musical Orchestration using Genetic Algorithms and a Spectro-Temporal Description of Musical Instruments", Proceedings of EvoMUSART Part II, Istanbul, Turkey, 2010.

• Esling Philippe, Agon Carlos "Time series analysis, sound mixtures and orchestration", Presentation in CNRS Japanese-French Laboratory of Informatics, Tokyo University, 2010.
• Esling Philippe, Agon Carlos “Composer les mélanges sonores avec les maquettes spectrales”, Journées d'Informatique Musicale, Rennes, France, 2010.
• Esling Philippe, Agon Carlos “Orchestration and Sound Mixtures”, Journées Jeunes Chercheurs en Acoustique Audition et Signal, Marseille, 2009.

• Robotics programming project - LI260 - Université Paris 6 - 54H
• From chipset to web - LI105 - Université Paris 6 - 10H
• Introduction to Object programming (JAVA) - LXTMI - Université Paris 6 - 32H
• Programming for electronics - LIE01 - Université Paris 6 - 32H

• Jordan Gerber : 2 Months internship - "Multiobjective techniques for sound retrieval"
• Pierre Lecomte : 3 Months internship - "Perceptual effects related to sound mixtures"