Information Technology Reference
In-Depth Information
Fig. 12.14
The musical rendering of the new FT that was generated by mutating the dir musicode
from FT1 and FT2
Mutating other musicodes (melody interval, event duration, note pitch, etc.)
would yield more variation. Mutations are possible across all musicodes in a similar
manner, with the only exception being mutations in modality. These are accom-
plished by a process of transformation whereby the intervals between successive
absolute pitches in the given FTs are forced to conform to preset intervals for major,
minor or diminished modes.
Finally, the new FT is rendered into a musical measure (Fig.
12.14
) and saved
into a MIDI
le.
The above example only showed the generation of a single measure. For longer
musical sequences, further FTs are generated by using the next FT in memory as the
source FT and mutating it with a target FT that again is selected according to the
value of the variable
of all other FTs stored in memory.
ω
12.7.3 Transformative Phase
The transformative phase comprises a number of transformation algorithms that
modify a given musical sequence, three of which will be explained in this section.
Although there are some differences in the speci
c processing undertaken by
each algorithm, the basic signal
flow is quite similar for all of them. The generated
input signal is modi
ed towards values given by one of the transformation algo-
rithms. With most of the transformation algorithms, the amount of modi
cation is
scaled according to the fMRI data. The fMRI data, or more speci
cally the data
extrapolated from the fMRI scans by ICA analysis, are referred to as the
fMRI_index. These data are provided to MusEng on a ten-point scale with values
between 0 and 9. In order to use the fMRI index as a control signal (CS) for the
transformation algorithms, MusEng
first scales the data to a range between 0.1 and
1.0. The system applies the following simple scaling process to the value of the
fMRI_index (Eq.
12.2
).
CS
¼
fMRI index
þ
1
ð
Þ
0
:
1
ð
12
:
2
Þ
f
g
A difference value
d
between the input and the transformed musicodes is also
calculated. This difference is then multiplied by the CS to give a
final scaled
modi
er value (SMV). The SMV is summed with the input signal to directly
transform the output. This gives a degree of fMRI-controlled variability in each
transformation: a high fMRI_index value will result in larger transformations to the
music, whereas a low fMRI_index value will result in smaller transformations.
