Atau Tanaka's BioMuse system is interesting. In this video he controls wind type sounds by moving his arms and hands. A simplified version of this with Wiimotes would not be too difficult to design.
https://vimeo.com/2483259
Tuesday, 9 July 2013
Wednesday, 3 July 2013
Wii Footsteps Update
This is the same patch as mentioned in the previous post but with an additional ability to adjust sounds dependent on speed.
A wiimote attached to the leg of the user measures their walking speed, when this reaches a certain value (currently a MIDI CC value of 30 but this needs to be adjusted) the MIDI channel is switched from 1 to 2 to control another sampler on MIDI channel 2 (it may be better to just add 12 to the MIDI value and store the running samples in the same sampler instance in Ableton).
This is a good improvement on the previous patch although sticking a wiimote in a sock is hardly the best way to do it.
Footstep samples for specific surfaces at different walking/ running speeds now need to be found to demonstrate this properly.
Still need to find a way to adjust velocity effectively (does a hard footstep have more high or low frequency content than a soft footstep? if so adjustment of high or low pass filters relative to the velocity info from the balance board could be worth experimenting with).
PD patch (new section highlighted blue).
A wiimote attached to the leg of the user measures their walking speed, when this reaches a certain value (currently a MIDI CC value of 30 but this needs to be adjusted) the MIDI channel is switched from 1 to 2 to control another sampler on MIDI channel 2 (it may be better to just add 12 to the MIDI value and store the running samples in the same sampler instance in Ableton).
This is a good improvement on the previous patch although sticking a wiimote in a sock is hardly the best way to do it.
Footstep samples for specific surfaces at different walking/ running speeds now need to be found to demonstrate this properly.
Still need to find a way to adjust velocity effectively (does a hard footstep have more high or low frequency content than a soft footstep? if so adjustment of high or low pass filters relative to the velocity info from the balance board could be worth experimenting with).
PD patch (new section highlighted blue).
Wii Fit Footsteps
In this configuration the wii fit board triggers recordings of footsteps in Ableton Live via Pure Data.
12 footstep samples (taken from one wav from the sony sound effects library) were loaded into Live's drum rack sampler.
Each footstep on the wii fit board triggers one of the 12 samples at random. At the moment there is a chance that the same sample will be triggered twice in a row, this is less than ideal as that could sound unnatural, I'm looking into ways to rectify that.
Although the PD patch is sending velocity information it is currently inadequate. Footsteps sampled at different velocities may be required but I'm not sure they are attainable. Sorting the samples into perceived velocity order and then triggering them based on the force applied to the wii fit board may work but I doubt that the board
Andy Farnell discusses the superioriy procedural audio over recorded audio for footsteps in Designing Sound. Although his argument is probably true this an attempt at a compromise. Working with audio samples does not take into account the different biomechanical movements made by the foot when moving at different speeds (striking with the heel or ball of the foot when running or walking). I cannot think of a way to rectify this but the ability to input speed information with a wiimote (either by a button press or by pitch or yaw info). [An idea - would attaching a wiimote or nunchuk to the users calf/ shin be able to determine speed information via the accelerometers and then switch to a running/ walking bank of samples when appropriate?]
Screenshot of the PD patch -
12 footstep samples (taken from one wav from the sony sound effects library) were loaded into Live's drum rack sampler.
Each footstep on the wii fit board triggers one of the 12 samples at random. At the moment there is a chance that the same sample will be triggered twice in a row, this is less than ideal as that could sound unnatural, I'm looking into ways to rectify that.
Although the PD patch is sending velocity information it is currently inadequate. Footsteps sampled at different velocities may be required but I'm not sure they are attainable. Sorting the samples into perceived velocity order and then triggering them based on the force applied to the wii fit board may work but I doubt that the board
Andy Farnell discusses the superioriy procedural audio over recorded audio for footsteps in Designing Sound. Although his argument is probably true this an attempt at a compromise. Working with audio samples does not take into account the different biomechanical movements made by the foot when moving at different speeds (striking with the heel or ball of the foot when running or walking). I cannot think of a way to rectify this but the ability to input speed information with a wiimote (either by a button press or by pitch or yaw info). [An idea - would attaching a wiimote or nunchuk to the users calf/ shin be able to determine speed information via the accelerometers and then switch to a running/ walking bank of samples when appropriate?]
Screenshot of the PD patch -
I'm finding that the possibilities of the wii equipment are more extensive than I first thought. I'm wondering if this should take priority over the Arduino experiments.
Subscribe to:
Posts (Atom)