Not, currently.
The second one. You’d just be scaling down the note values to the range you define with the attentuator. The relationship between steps would remain as they were set on the device. The device is intended to play notes. You’ll want to stick to something like the MengQi Voltage Memory or Pressure Points if you want arbitrary voltages. Alternatively, you could write your own firmware for TSNM, the source code is available, but that would be a lot of work for what I imagine you are trying to do.