Playing with art and technology

By Nina E. Tveter

Up to now, the transverse flute has always sounded like a trans-verse flute. In the not too distant future, transverse flutes will also be capable of sounding like a guitar, a didgeridoo or rush-hour traffic in Paris. As part of her doctoral research work, Sølvi Ystad has developed a sound simulator which makes it possible to play any sound on a traditional instrument. As Ystad has played the transverse flute for 20 years, she chose to develop the technique on this instrument.

The hybrid flute

Nowadays it is quite common to use synthesizers in order to create new sounds or to imitate traditional sounds. Ystad says that such instruments are no longer of such interest to professional musicians, who want to extend the range of new sounds which they can create, at the same time as they continue with the playing techniques which they have already acquired.

Sølvi Ystad made a digital transverse flute by blocking off the bottom of a traditional transverse flute, connecting sensors to the finger valves and a microphone to the mouthpiece.

­ You cannot interpret music on a keyboard in the same way as you do with a flute. The air pressure, the position of the lips, and the fingering, are crucial elements for the sound ­ and they are impossible to replicate on a keyboard instrument.

For this reason Ystad created a digital transverse flute by blocking off the bottom of the traditional transverse flute, and connecting sensors to the valves and a microphone to the mouthpiece. She linked up the technology on the flute to a computer program she had designed herself. As a result, flautists can now gain access to a virtual world offering innumerable sound effects. For example, you can play other instruments such as the guitar or the piano on the flute, or you can make the flute sound as if it were 30 metres long. Or you could make a note start off sounding like a flute and then transform itself into the sound of a clarinet...

Ready to face the world

Normally, electronic sounds are created when someone tries to recreate a known sound in a completely synthetic manner. Ystad chose to change this approach; she looked at the phenomena which took place when she played the traditional transverse flute as a physicist - and then she tried to imitate the physics which lay behind the sound.

This was a very challenging project, as the physics of wind instruments is extremely complex (for example: the column of air which is blown into the mouthpiece creates turbulence). Her research has also been linked to psycho acoustics, as she used various listening criteria in order to isolate which phenomena are the important ones.

Ystad has also been in contact with French composers who are willing to compose music for this 'new' instrument.

Norway and Norheim

­ France and Sweden are the two European countries where you find the best research environments within the field of musical informatics. This is where electronic music is being composed, and concerts involving this kind of music are held regularly in Stockholm, Paris and other major cities in these countries. In Norway, electronic music stops with Norheim. Apart from a growing interest in techno musical circles, in Norway there is hardly any involvement with this type of music or research into the technology of music, explains Ystad.

For this reason Ystad started collaborating more and more with sound researchers at the University of Marseilles. She has just been awarded a doctoral degree in signal modelling from both the Université Aix-Marseille II and NTNU. This work was founded by the Research Council of Norway.

Ystad's supervisor, university lecturer Jan Tro at NTNU's Department of Telematics, thinks that she has done a good job in extending a mathematical basis for the physical modelling of musical instruments. Ystad is also the first person at NTNU to combine different methods in sound modelling. Jan Tro thinks that it is very likely that her work will help to improve Norway's international profile in the field of music technology.

French car industry

Her doctoral work gave birth to more than a musical instrument. The re-synthesis (or recreation) of musical sounds has also resulted in an analytical tool which Ystad can make use of in completely different areas, such as that of noise pollution. Working with a national research centre in France, Ystad is helping the French car industry to eliminate the sound frequencies which bother drivers and passengers in new car models.

The French telephone industry also appears to be interested in her competence in the field of sound:

­ In France there is a problem with telephone poles which need to be replaced because of rot. At the moment we have people with a specially trained ear who walk around knocking on the poles in order to single out the bad ones. Even if they are good at 'separating the wheat from the chaff', they do make mistakes. This means that many good poles are being replaced unnecessarily.

In future, Ystad's techniques of analysis will enable us to use a digital instrument to function as an 'ear'; so that anyone can perform the task of finding the decaying telephone poles - and with optimal results.

Sounds from the numeric
transverse flute can be heard on:
www.oslo.sintef.no/gemini/1999-04/sound/

Contact: Sølvi Ystad
E-mail: Ystad@alphalma.cnrs-mrs.fr