On Wed, 19 Oct 2016 17:31:42 +0200 Robin Gareus <robin@xxxxxxxxxx> wrote: > For deep learning, the analysis tool itself would be the DAW in this > case which has context: All tracks, random data access, and it can > control the plugins. This is the question I have raised, which was expressed by where to store data but extends naturally to where would the decision-making entity reside. Having distributed knowledge in which every part contributes to the decision would push this kind of endeavour way out of the audio domain. On the other hand, embedding this in a DAW imposes not only maintenance issues (updating the DAW for updates in the decision making process) but also restrictions regarding portability, which would limit wide acceptance. > There was an endeavor to standardize a very basic part: Allow a plugin > to analyze a complete track and write its own automation data: > http://lists.lv2plug.in/pipermail/devel-lv2plug.in/2016-February/001571.html Thanks, this looks quite interesting. And is out of the user domain, into development. Using LV2 of course makes away with the notion of portability across many well-known DAWs. But it might be the only option in open source. Could be that none of the available licences can apply when developing open source using VST, I don't know. > Something like this would be a very basic building block: the plugin > knows everything about itself, can do analysis and then "control" > itself. Even without fancy inter-plugin communication this could go a > long way: gates, adaptive de-esser, pitch-correction, noise > cancellation, heck even a channelstip plugin. Wouldn't some of the recent Harrison plugins adopt something vaguely similar to this ? > There are working prototype implementations (host & example plugin) > but so far it came to nothing. Go figure. Well, if the characteristics of the discussion here is a sign... :) > The actual use-cases were eventually implemented as VAMP-plugin > (Polarity optimizer) and plugin-internally (XT-TG Tom Gate). Ah ! Any effort done towards was is sketched here would have to be done using a solid base of functional plugins. There is absolutely no use in developing plugins on top of that effort. Harrsion, Calf, for instance. Thanks for the link, I'll take time to go through the technical discussion. _______________________________________________ Linux-audio-user mailing list Linux-audio-user@xxxxxxxxxxxxxxxxxxxx http://lists.linuxaudio.org/listinfo/linux-audio-user
