Squencing clip text and seding out via OSC…
What I am envisioning…
I would like to create a clip on an Ableton track that has some
text associated with it. This text would basically be score commands
that would be picked up by M4L and sent out via OSC to various
external sound tools.
Is it possible to associate text with a clip and access in in M4L?
Maybe you could store your text commands in a coll and trig them using midi notes in your clips ?
The point is to store the text in ableton clips, so I can move them
around on the time line. I really don’t want to have to manage a
midi note -> message mapping.
You know about ClyphX, right?
Well ClyphX looks very impressive. But I don’t see any
commands that route data out OSC. Did I miss something?
You can assign Info Text to any clip in Live but for some reason this attribute is not available in the Live Object Model. Cool idea though but I think you’re SOL. This would be a really simple enhancement for Ableton to make, so maybe it’ll be supported in some future release
that’s the reason why I suggested using coll objects and midi mapping…
The short answer is "yes", that’s possible. Sorry for the brevity earlier…my point was just that it can be done (if you can do it in Python, you can probably do it in m4l). I just heard of ClyphX the other day, and thought it might contain some clues. I wouldn’t want to do what you’re talking about without js ,but then I don’t really do much in Max without js, so that’s not really saying anything.
I think you will, however, find it easier to use conventional methods (per Stephane, or some other means) than to learn and implement m4l to do what you want. Either way is going to require a fair amount of programming if you’re not starting from some previously developed framework. Not to mention you have to parse the OSC on the other end of things…
If you decide you want to do it, I’m glad to help however I can. I’ve thought about implementing something similar in my own work lately.
Midi mapping could be done quite elegant using program changes as they are associated directly with clips.
I get around this by serializing metadata into json dictionaries which I set as the clip names.