Teensy3 String Machine

Not open for further replies.


Well-known member
It's working good enough to share so here it is: https://github.com/quarterturn/teensy3-string-machine

This is a 32-voice 'string machine' meant for the pjrc.com teensy 3.6 and the blackaddr audio interface. It depends on the Teensy Audio Ensemble Chorus library found here: https://github.com/quarterturn/teensy3-ensemble-chorus.

I think it sounds pretty nice. 32 voices is enough for long release settings without audible note stealing. CPU utilization is about 30% on a Teensy 3.6, so it will probably run on a lesser Teensy 3.

The ensemble and phaser are fixed, as for me the represent the sweet-spots of their setting for what I want. You can control the dry/wet mix of the phaser. If there's a big call for it, I'll make them configurable.

Check the readme.md on github for more info.
demo video

Here it is: https://youtu.be/yDEagM8vqhU

Audio is just through the phone mic, as at the moment I don't have a convenient stereo-in capture setup. The ensemble effect is stereo and would sound a lot better as such. Everything you hear is from the synth code, including the phaser. I tried to include some long release settings to demonstrate the 32-voice poly. Unfortunately you can hear that the envelope function is linear; release tails end abruptly and sound rather unnatural. Exponential curves would be much better.

My Macbook dev environment is unfortunately broken at the moment thanks to upgrading to Catalina.
Not open for further replies.