Yes, that was the project I remembered. Thanks for the hint, the Airbow looks interesting!
1 Like
Just throwing this in. I’ve been working on this for a couple of years and now gigging with it regularly around Europe. Granted, the hardware is not of my own design (a Leap Motion), but everything stands or falls with the software, in this case a Maxforlive device of my own design. It is everything I want. Gone are days of statically staring at a laptop. In principle being able to control a limitless amount of sound parameters and the possibility to really perform. I’ve done many pieces with it and quite a few are on my YT channel. I hope it inspires!
https://youtu.be/4p82LIZN1DA
2 Likes
Although I’ve been aware of the Bela Trill range of sensors for a while, I’ve only just realised that the Trill Flex is based around the idea of designing your own custom cap-touch sensors with up to 30 contacts. I reckon a couple of these might work for a project I’ve been wondering about for a while, which is (re)building a faulty Misa Tri-Bass, so I’ll give that a go for the fretboard.
If I use Teensy to drive this, should it also be able to handle a simple interactive touchscreen, or would I be better off using RPi for that…?
I was looking at the latest video for the Soma Terra
…and thinking that a chunk of wood and a Trill Flex might be a way to approach a similar type of controller (without the synthesis side), although their cap-touch sensors seem to detect a variable response rather than the usual on/off detection.
(Someone pointed out the resemblance to the rare Latronic Notron sequencer which may be one of its design inspirations.)
3 Likes
Here’s something I’m working on. I was inspired by the Duduk to make a simple mono instrument with notes belonging to one scale. Trying to find the minimum viable expressive synth.
It uses a flexure to provide the ability to press down for emphasis and wiggle for vibrato.
3 Likes
Here’s a video of the prototype in action. The patch is a sawtooth wave going into a lowpass filter. The amp envelope is basically the gate signal from the touch sensors, with a bit of attack and release to mitigate clicks. The pressure signal modulates the low pass filter, and I also added a decay envelope to “pluck” the filter when a new note is started.
Edit: it works using a flexure and Hall effect sensors to measure the motion
5 Likes
cool stuff @TigerBalm … are you using a daisy?
also how do you find 3d printing for something of this size…
what do you use?
more generally, if you have time, it’d be create to create a topic on your adventures in this area, with some details, and thoughts … but also like the above, discuss your experience with things like 3d printers.
edit : Ive retitled the topic, to reflect its not just about midi controllers.
also, perhaps we can break this topic out a bit, into how we might cover different aspects of it.
i.e. use this section, to help one another start building these kind of projects
Ive created a new topic about housing/form factor , as this is something Im desperately struggling with.
But, if you want, feel free to create new topics, about other aspects … e.g. DSP , electronics.
(or even more specific areas in these), Id enjoy contributing to these where I can.
1 Like
That’s great! If it’s done entirely with Daisy, it’s amazing.
1 Like
Thanks for nice comments technobear and @BJG145
Yup it’s all Daisy. I’ll start a thread on this project soon when I have a bit more details to share.
I printed it on a prusa i3 mk3s+. My background is in mechanical engineering so the enclosure stuff comes naturally to me but the electronics and programming I take one tiny step at a time.
3 Likes
Awesome. The only thing I could find to get any synth sounds out of my Daisy Pod was the OscPocket VASynth port, but I don’t remember it sounding as good as that. Pollen8 is long gone on new chips. The way you’ve handled expression on both the hardware and software is also impressive.
This is a good advert for Daisy; it ain’t an easy option.
1 Like
This weekend I recorded a new demo for the “Eviola”, a prototype bow-operated MIDI controller using EVI fingering developed in collaboration with John Dingley (Digigurdy).
It uses four cap-touch sensors on the bow (three for the “valves” and one for vibrato), plus four on the body of the instrument for half-octaves. It generates MIDI breath pressure via a slotted bow and an optical sensor which can be used to play instruments like SWAM.
2 Likes
Great project! So, you do vibrato on the bow? Is the playing board discrete or ribbon-style continuous?
1 Like
Thanks! One of the buttons on the bow is vibrato on/off. The fingering is chromatic over a 3 octave range; the cap touch is just on/off.
2 Likes
Question for thetechnobear. Earlier in this thread, back in Jan 22 you provided an excellent summary of your views on the state of embedded sound engines, covering Teensy, Daisy, RPi and Bela.
18 months on, I was wondering if you might be able to offer an update. Have your thoughts changed…? Are there new products to consider…? Having taken the plunge with PCB design for MIDI controllers via Fiverr and JLCPCB, I’m again considering how I might add a simple (or even complex) embedded synth engine to these devices. Initially I’m thinking about the “Screech Owl” ribbon controller mentioned above that I want to build, along the lines of the discontinued Eowave Persephone.
1 Like
interesting question…
its been 3 years since I made that video, so you would think a lot would have changed in 3 years, but I dont think it has really.
Id have to look at release dates, but I think the main change is perhaps more offerings from raspberry pi, particularly pi zero and pico - they perhaps close the gap between the teensy to pi.
daisy, is definitely now more established, and so I think easier to get started with.
and they have patch.init which is a nice off the shelf eurorack offering, and patch submodule to create your own eurorack modules.
bela, their new gliss module is an interesting option in the eurorack space.
… and trill is a very good way to get started with sensors with minimal experience (and works with other platforms as its just i2c)
befaco have lich, but thats pretty much the same as owl
(they redesigned, and took it over afaik)
oh, and axolotl, is unfortunately no longer viable
(cannot remember if I mentioned it in the above video)
are there any new products in this area you can think of?
(Im sure Ive forgotten one or two… and my list was never exhaustive)
in many ways, I think, the offerings I mentioned really already covered most use-cases, so Im not sure we will see more.
I think the main change was daisy wanting to ‘standardise’ things a bit esp in the eurorack arena. its main advantage now, Id say is you could create a desktop product with it, then easily move it to a eurorack offering…
however, if its just for you, for diy, many won’t need this - they already have a form factor in mind?!
but also its popularity will mean, that perhaps others have created support for things like sensors and displays, something that was always useful in the Arduino space.
3 Likes
“I think the main change is perhaps more offerings from raspberry pi, particularly pi zero and pico”
Thanks for the update. Yes, I just came across Pico in the last couple of days. I was looking at this MIDI board earlier.
1 Like
I have a vague interest in Elk Audio OS, especially since the NINA synth uses it. I cannot afford a NINA at the moment but its possible that hacking that synth might be in my future as a starting point, if they follow through on their promise to allow users to hack it.
2 Likes
There is potential. As far as I know, only the Synthstrom Audible Deluge has ever made the transition from proprietary to an open-source license in the music industry, so it would also be huge news if Melbourne Instruments follows through a similar transition too.
…oh, interesting. Need to explore that… 
1 Like
It will be misleading advertising if they dont, since the features section on their own website still says “Hackable Open Source software built on a powerful Raspberry Pi 4 running Elk Audio OS.”
All the same I wont make any assumptions until they actually facilitate this side of things.
1 Like