Brainstorming Session:

1. data look like FFT analyses - so use FFT and Convolution bins directly? 2. or alternatively, a multi-filter resonator with e.g. 150 bands, maybe detuned from a harmonic series?

Wish list: get real impulse responses of the measurements, i.e. rawer data, for audification

1. idea: do 1-2 seconds signature for each file by parameter mapping, e.g. 15-20 dimensions; and try 2 beacons binaurally.

Experimentation

look at data ... all channels 1 seems saturated in the upper half, so:

take the undistorted part of channel1, downsample it to e.g. 50 zones, make these 50 resonators -> these will ring differently for the different materials when excited

Evaluation

two sessions are much better for deeper discussion and more interesting results. Even so, more time would be desirable. The sound signatures made were easy to distinugish, so in principle, this works.

What next?

We would like to implement signatures of more tahn one channel to increase reliability of properties tracking; e.g. for materials production monitoring, this could be a useful application.

It would also be interesting to try a nonlinear complex sound generator (e.g. a feedback FM algorithm) and control its inputs from the data with ca. 20-30 parameters; this approach is not yet well understood, but could be intersting as sonification research.