MIDI transpose with rotating voices

[mattmarantz]

Hello all, I'm trying to figure out a way to do this in Usine HH3: https://www.facebook.com/vlade.guigni/p ... 2805552429

Basically, his Matrix 12 is set so that each voice is detuned to a different note which in total makes up a musical phrase. He can play the whole phrase by manually retriggering the same note on the keyboard for the duration of total notes in the phrase, or where it really starts to get interesting is when he plays improvised linear lines and the rotational transposition process creates some really interesting note ouputs which a normal musically trained human wouldn't come up with on their own! Sounds so cool.

I want to do this in Usine, but I don't know where to start... Is there slmething in the add-ons secrion someone already made that can do this, or can someone please point me in the right direction to get started with building a simple add-on that would do this?

Thanks!

Matt

[x.iso]

I don't really get the idea, what do you mean each voice detuned to a different note? different from what?
There is a midi expand voices script, which divides incoming notes to different voices (you can edit the script to change number of output voices) and then you can process those notes independently, so that's probably a start for what you want to do. There are also options for the way voices are counted, but I'd change the script so that incoming value could determine which voice to output as well.

[mattmarantz]

I don't really get the idea, what do you mean each voice detuned to a different note? different from what?
There is a midi expand voices script, which divides incoming notes to different voices (you can edit the script to change number of output voices) and then you can process those notes independently, so that's probably a start for what you want to do. There are also options for the way voices are counted, but I'd change the script so that incoming value could determine which voice to output as well.

Well as far as clarifying what the idea is, was the link I posted viewable for you? I was not sure whether or not the link is private or public. It's my friend Santiago Bosch who is an amazing keyboardist in Boston and he has an Oberheim Matrix 12. The Matrix 12 had a unique feature where you could tune each of the individual voices to a different note. So, the voices (when played independently, one after another) would trigger whatever notes you tuned the original voices to in order one by one.

So, let's say you tuned the 1st voice to regular C, then the next voice to a wholestep up D, then the next voice a wholestep up from there to E.

That way, when you played a C 3 times in a row, it would play C->D->E instead of just 3 C's.

You could continue that operation based on your programming all the way up to 12 notes which is the maximum amount of voices the Matrix 12 had.

The use for this is coming up with interesting harmonic passages during linear improvisation. If you are able to listen to Santiago's video, it's pretty clear how awesome this can sound.

Like, let's say you program in an interesting musical phrase of 8 notes into the script, and then you also vary the input notes too, like during an improvised solo. The results would be super interesting and abnormal. It would allow people to play intervals and musical passages triggered by more traditional improvisational language that would not normally occur to them (well, would not normally occur to me, anyway.) Basically, something to transpose your incoming midi notes to something other than what you are playing, based on intervals that you are allowed to program that rotate consecutively in order until they get back to the beginning of the rotation and repeat infinitely until you stop playing.

My friend is working on a .vst version of this but I want to do it in Usine. Keeps everything more compatible and streamlined.

Does that make sense?

[x.iso]

I've seen the video, but wasn't sure if that was the idea.
So, as I said you patch "midi expand voices" to midi in, then change the script if you need different amount of voices and patch midi transform to each output voice and transpose each voice to desired value or alternatively change the scale for each midi transform., then patch all midi transform modules to single midi out. that's it.

[x.iso]

I'll upload later a patch that I'm putting together, that will do what you described and a bit more.

[mattmarantz]

I'll upload later a patch that I'm putting together, that will do what you described and a bit more.

Thanks! I actually used your instructions and I was totally able to make it work as desired! Thanks so much. That was fun. What an amazing program, I didn't realize quite how deep it is!

Everything worked except MIDI Breath (from my EWI wind instrument, my primary controller) so I solved this by creating a 2nd MIDI bus with direct control from the EWI but filtered out the notes, and added the 2nd bus to the input of the instrument as well as the voice-expanded-processed MIDI bus.

Why is my patch, built exactly as you described, filtering out the expressive MIDI data like MIDI Breath and Aftertouch? Is there a way around that? That's the last part of the equation for me, everything else is solved!

I'd love to see what kind of patch you come up with, though, if you want to share.

Cheers.

[x.iso]

the reason why midi breath might not work is probably due to midi signal getting switched to different channels (voices) in the script all the time. I haven't thought about that and I think I might need to workaround this in my midireseq patch.

as for why your workaround didn't work, I think if you have several midi outputs in a patch that is not a sub-patch, then it will output only first midi out to the next patch or rack output. So you just have to add midi filter into patch, connect it in parallel to midi expand, filter notes out and patch output to same midi out.