Study for a Vocoder
I recently concluded a residency at LEMUR with a performance Study for a Vocoder. For quite a while I've been wanting to build a few vocoder-like sculptures/musical-instruments... and this is the first step towards the first of them.
It's built out of a ink-jet printer and a slide-whistle. It replicates whatever sounds are spoken/sung into a microphone. It can also play pre-recorded music ala a player piano. Its main short coming is being a bit slow in changing notes, but a more powerful stepper motor should help improve its response time.
Labels: acoustic, interactive, music, physical computing, sound
posted by Ethan at
4:39 PM
5 Comments:
You got it done! Uber-cool!
Now we just need about five more copies of it so that we can start doing real vocoder stuff.
Is the airflow modulatable, or just bang-bang?
Just bang-bang for now :)
Yeah, more pipes! I think I'll need 3 of various sizes to get a good tonal range.
I played around with Pure Data's Fourier analysis tools to see how many simultaneous notes are needed to replicate vocal tones... it seemed like no more than 3, and that really even 2 would do fine. But perhaps that's really showing a limitation with the Fourier tool (~fiddle) than what needs to be done.
Ha! Interesting timing! Got a Stamp-controlled mini calliope working today; only took a year to get the bugs out, hehe. I suspect it'll take another year to make it as elegant as your slide whistle tho... ;-)
www.youtube.com/watch?v=Ifq7Z9ZSBl0
Excellent work. I ordered a couple slide whistles last week to attempt build of a midi slide whistle - now i see that someone has done this. Do you mind posting a couple technical details on the mechanical design? For instance, the type of air valve and air pump or fan?
Congratulations on a nice implementation.
Hi Tom,
I'm currently using a fairly large shop-type air compressor with the pressure brought down to about 10-15 psi. Higher pressure gives me unwanted overtones. Eventually I'd like to switch to using a air chest, which is the solution used by pipe organs.
I experimented with two different solenoid valves for controlling airflow. They are relatively expensive (around $30) and louder than I like (you can hear the clicking on the video). I think it's worth looking into Steamboat Ed's solution of water sprinkler parts to see if it is quieter and/or cheaper. (Alternatively, putting some sound insulation around the solenoid is probably called for).
Of the two valves I tried out, the marginally quieter one came from www.ValveStore.com. Here's the details:
42703 Solenoid Valve 2-Way NC 1/8", 24/DC 42703 Solenoid Valve 2-Way NC 1/8", 24/DC, $31.95 with shipping.
The other valve was from www.stcvalve.com. 2V025-1/4-2-D, 1/4" NPT Direct Acting 2 Way Solenoid Valve (24VDC DIN). It was $26.44 with shipping. As mentioned, this one was slightly louder.
Post a Comment
Links to this post:
Create a Link
<< Home