Revealing the harmonic structures of your brain
The Biotuner is a python toolbox that incorporates tools from biological signal processing and musical computation to transform biosignals into microtonal musical structures (see the Github). Hence it extracts musically relevant information from biosignals.
The toolbox works by first extracting spectral peaks from a time series. Then, tuning systems are derived based on a set of music computation algorithms, such as the dissonance curve (William Sethares) and the harmonic entropy. More than deriving tunings, the toolbox provides different harmonicity metrics, as well as rhythmic structures and cross-frequency coupling indicators.
The computed tunings can be use in realtime for musical compositions, as well as for visual compositions.
For musical composition, tunings can be exported in .scl format and therefore subsequently imported in Virtual Studio Technology (VST) that supports microtonal scales. Depending on the type of scale construction algorithms and peaks extraction methods that are used, the biotuner can also be incorporated into realtime neurofeedback systems (see: https://github.com/AntoineBellemare/EEG_m4l/)
The following image represents the dissonance curves associated with time series of multiple EEG sensors. The resulting tuning has steps at the frequency ratios indicated on the x-axis.
The toolbox provides different visualization tools. First, the consonance matrix, which represents the degree of consonance for each pairs of notes within a tuning.
More aesthetic visual representations can be achieved, with the Lissajous curves, which are geometric representations allowing an intuitive visual impression of notes harmonicity.
The range of visible frequencies (400–790 terahertz) spans approximatively one octave, which is also the span of octave-based musical tunings span. Consequently, a representative color palette can be constructed from each tuning, with colors corresponding to the frequency ratios. The hue information of each color is derived by multiplying the frequency of a fundamental color by each ratio of the tuning. The saturation is derived from the harmonicity of each tuning interval within the whole tuning.
A more saturated palette means that the tuning is inherently more consonant.
Color palettes can also be derived directly from a series of spectral peaks. EEG frequency peaks are converted into frequency values within the visible spectrum. This is simply done by multiplying the frequency by 2 (octave) iteratively until the it falls within the visible spectrum. Luminance (luminous intensity) is derived from the amplitude of each frequency peak, and saturation from the distance between a peak and the fundamental frequency.
Improvisations with the Biotuner.
These snippets aim to show some examples of harmonic progressions that can emerge from biotunings. New tunings equate with new harmonic horizons.
Excerpt 1
Excerpt 2
Excerpt 3
Excerpt 4
