After four years with our laboratory, Roberto Bresin has defended his
thesis for a doctoral degree, entitled "Virtual virtuosity, studies in
automatic music performance".  To cite the beginning of its abstract:

"This dissertation presents research in the field of automatic music
performance with a special focus on piano.

"A system is proposed for automatic music performance, based on
artificial neural networks (ANNs).  A complex, ecological-predictive
ANN was designed that _listens_ to the last played note, _predicts_
the performance of the next note, _looks_ three notes ahead in the
score, and plays the current tone.  This system was able to learn a
professional pianist's performance style at the structural micro-level.
In a listening test, performances by the ANN were judged clearly better
than deadpan performances and slightly better than performances
obtained with generative rules."

The main text of the thesis summarizes its component six academic
papers published in Journal of New Music Research and Computer Music
Journal among others, and JAPER and PANN, two Java applets for music
performance, included in the CD-ROM MidiShare: Operating System for
Musical Applications, see http://www.grame.fr/MidiShare

The abstract as well as the thesis itself (PDF file, 275 kb, 42 pages,
excluding the published papers and software) can be downloaded from
http://www.speech.kth.se/music/publications/thesisrb/

Johan Liljencrants

 [ A practical use someday of this technique might be to add expression
 [ coding to a non-expression piano roll.  The computer would analyze
 [ a fine pianist's performance from a well-coded reproducing piano roll
 [ or MIDI recording, and then apply this pianist's "performance rules"
 [ to a classic 88-note (or 65-note) piano roll.  The result would be
 [ a MIDI file with hammer velocities (expression) which could be played
 [ on a solenoid piano.
 [
 [ Prof. Liljencrants teaches speech communication and electroacoustics
 [ at the Royal Institute of Technology, Stockholm, Sweden.  See his
 [ professional home page at http://www.speech.kth.se/~johan/index.html
 [
 [ -- Robbie