Here’s a piece that I’ve been working on for a while… sort of a class project using my “midiControl-harmonicRotator” Max-patch (I’ll post it if there’s interest).
Anyway this performance is by my housemates Becca (Mezzo-Soprano) and Joel (Yamaha WX-5 wind-controller).
Perhaps I’ll post a score and the patch later.
I forgot to apologize and warn about occasional clipping in the file.
Seriously interesting. How does it work?
Good question! First, Here is the patch and score, as promised.
I have trouble articulating what it does, but let me try again: The program takes an input from an audio source and a midi source. The audio source is then output as the original signal, the spectrogram of the frequencies shifted down by half of the midi-input note, and the spectrogram of the frequencies shifted up by the midi note.
So that’s a little hard to imagine so for example if the MIDI note and the audio note are the same, the result should sound like that note in three octaves. So it sounds nice and clean. But when, for example, the notes are a whole-step apart, the result would be a strange, out-of-tune three note chord.
What I do a lot in the piece is put the note far apart which results in one of the three notes leaving the audible range, so we only hear two notes.
Hey man,
I like this- sounds like a glorified ring modulator. I also like that there are two musicians involved, but you’re really only hearing the sound of one musician.
Sym, I just listened to this, which as I told you I intended to do a long time ago. I like it. There’s one moment just pass half way, but I think before 66% (the mp3 player in the browser window didn’t show time) where it does this perfect little kind of embellishment thing. I think she’s holding a note and the MIDI’s moving. I dunno, but it’s awesome.
One idea I had (and I dunno if it’s possible or even worth the effort). This patch (and the notes written for it) is all about relationships: to notes, to the performers, to man and machine, etc. What if you included some kind of timbral relationship also? Like when she sings different vowel sounds, and in each one of those is a different set of relative frequencies right (I think I remember that from that Physics/Music class we took). So what if those relationships caused the newly generated tone(s) to change their timbre? Simple example: from square to saw. With the right tweaking and subtlety, I think that might be very sweet indeed.
Cool man. I’m guessing you’re thinking about what’s going on at about the 3-minute mark.
I’m glad that the “relationships” part of the piece is obvious… I like your timbral idea, but I think that’s another patch if not another piece. If you have a peek at the MAX/MSP end of things, you’ll see that the “wet” output are shifted spectra of the vocal input, so they are changing timbrally to match the voice’s vowels.
I’ll admit that it’s subtle though.
Subtle meaning imperceptible.