Information Technology Reference
In-Depth Information
fMRI still is impractical for this purpose: fMRI scanning is simply too expensive to
run, the equipment is not portable, and the health and safety implications for usage
outside strict laboratory conditions are
fiendishly burdensome. Moreover, fMRI
scanners produce noise during the scan, which makes it inconvenient for a musical
application. We are currently working on detecting in the EEG equivalent activa-
tions in auditory cortex as we detected in the fMRI scans.
In the meantime, I have been developing generative music systems suitable for
control with information representing cortical activations of tonal processing. I
teamed up with Torsten Anders, then a research fellow at ICCMR, to implement a
prototype that generates chords sequences automatically, in the style of the ones
used as stimuli for the tonality experiments (Miranda et al.
2008
).
We adopted a computational paradigm referred to as
constraint satisfaction
problem
to implement a generative music system that generates sequences of chord
progressions in real time (Anders and Miranda
2011
,
2010
). The input to the system
is a stream of pairs of hypothetic brain data, which controls higher-level aspects of
chord progressions. The
es whether a progression
should form a cadence, which clearly expresses a speci
first value of the pair speci
c key (cadence progres-
sion), or a chord sequence without any recognisable key (key-free progression).
Additionally, if the next progression is a cadence progression, then the key of the
cadence is speci
ed by the second value of the pair.
Each chord progression (Fig.
1.10
) consists of
n
major or minor chords (in the
example
n
= 16). Different compositional rules are applied to cadence and key-free
progressions. For instance, in the case of a cadence, the underlying harmonic
rhythm is slower than the actual chords (e.g. one harmony per bar), and all chords
must fall in a given major scale. The progression starts and ends in the tonic chord,
and intermediate root progressions are governed by Schoenberg
s rules for tonal
harmony (Schoenberg
1986
). For a key-free, atonal progression, the rules estab-
lished that all 12 chromatic pitch classes are used. For example, the roots of
consecutive chords must differ and the set of all roots in the progression must
express the chromatic total. In addition, melodic intervals must not exceed an
octave. A custom dynamic variable ordering scheme speeds up the search process
by visiting harmony variables (the root and whether it is major or minor), then the
pitches
'
finally the pitches themselves. The value ordering is
randomized, so the system always produces different results.
As it is, the generative system design assumes that subjects would be able to
produce the required control information in some way or another. In practice,
however, it is unlikely that subjects would learn to produce bilateral activations of
transverse temporal gyrus simply by imagining tonal progressions. The challenge
here is to establish effective ways to embed in a realistic system design the theo-
retical understanding of neural correlates of tonal processing and generative musical
algorithms. The research continues.
'
group (or classes) and
