CORDIS
EU research results

CORDIS

English EN

Directionally Optimised Representation of Musical Instruments

Objective

The DOREMI project consists of an experimental study of the directional characteristics of musical instruments in the context of their performance in a room. The overall objective is to develop and optimise a recording system that will allow a better representation of the directional characteristics musical instruments, which will be more accurate than the ones already existing. This is expected to be a significant step forward to achieve highly realistic room auralizations. Research on further developments of the system will be carried out in the field of virtual reality systems, moving sound sources, computer music and multiple-channel reproduction. The DOREMI project consists of an experimental study of the directional characteristics of musical instruments in the context of their performance in a room. The overall objective is to develop and optimise a recording system that will allow a better representation of the directional characteristics musical instruments, which will be more accurate than the ones already existing. This is expected to be a significant step forward to achieve highly realistic room auralizations. Research on further developments of the system will be carried out in the field of virtual reality systems, moving sound sources, computer music and multiple-channel reproduction.

OBJECTIVES
1. To develop and optimise a multi-channel recording method for musical instruments in an anechoic environment;
2. To study the behaviour of the radiation characteristics of musical instruments in a real performance situation;
3. To make multi-channel recordings of different kinds of musical instruments;
4. To optimise the spatial representation of musical instruments in room auralizations;
5. To study the application of the recording method for other purposes (virtual reality systems, multi-channel audio reproduction systems, musical ensembles, moving sources and computer music).

DESCRIPTION OF WORK
The work will consist of testing different kinds of set-ups of the recording system with different number of microphones and positions. Various simultaneous multi-channel recordings will then be done with musical instruments of diverse directivities. They will either be played by professional musicians or will have mechanical excitation attached to their bodies. The multi-channel recordings that result will afterwards be used as a reference for room auralizations where the differences in set-up will be compared according to the perceived spatial quality of the sound source and also the spatial sound distribution in the room. The representation of the source in the auralizations will be done with a number of virtual sources and positioning that corresponds to the one of the recordings. Considering the listening experiences and the directivity of the instruments obtained from the recordings, the system will be optimised and new possibilities will be tested until the best possible spatial representation of the source has been reached. When the recording system has been optimised, the same set-up will be used to study different possible of applications for multi-channel reproduction and virtual reality systems.
The multi-channel system was implemented and used for recordings and auralizations of musical instruments. Listening tests showed that the new method improved considerably the sound quality in terms of the naturalness of timbre. The directivity of a large number of musical instruments were measured using two different methods, and the results can be visualised by a specially developed computer program created for this purpose. The multi-channel method has also been successfully applied with a duo of instruments, and with a piano in a studio with non-anechoic conditions. As a futher result it has been proved that the multi-channel representation can be used for an alternative source representation (La Timée). It was also demonstrated that studio multi-channel recordings can be used as a basis to create a type of stereo recording mix that could bring an alternative timbre to the one done with traditional stereo techniques.
Leaflet | Map data © OpenStreetMap contributors, Credit: EC-GISCO, © EuroGeographics for the administrative boundaries

Coordinator

DANMARKS TEKNISKE UNIVERSITET

Address

Anker Engelundsvej 1, Bygning 101a
2800 Kgs. Lyngby

Denmark

Administrative Contact

Jens Holger RINDEL (Professor)

Participants (1)

Sort alphabetically

Expand all

INSTITUT DE RECHERCHE ET DE COORDINATION ACOUSTIQUE MUSIQUE - IRCAM

France

Project information

Grant agreement ID: IST-2001-39158

  • Start date

    1 October 2002

  • End date

    30 September 2003

Funded under:

FP5-IST

  • Overall budget:

    € 100 000

  • EU contribution

    € 100 000

Coordinated by:

DANMARKS TEKNISKE UNIVERSITET

Denmark