general
-
Hi guys, so I found this on GitHub and I wonder if it's possible to include this into HISE or JUCE. just to create a VSTi plugin ?
https://github.com/suno-ai/bark
If not is there another way that I can create a text-to-sample plugin that use AI ?
-
@ClawFORGE-Studio it's text to speech?
-
@d-healey Yes, but I wonder I can make a Text-To-Sample into HISE as well.
-
@d-healey Because I've been looking online for AI Stuff and I wondered if I can make a plugin using AI, for example I've found OpenAI, Bark by Suno AI and a few more ones and I wonder if it's possible to include it into HISE.
-
@ClawFORGE-Studio Well you can fetch the repo using HISE methods, download it to wherever, and then run the python process using the BackgroundTask class, then dynamically load the output into HISE.
You'll have to either wrap the whole thing along with python into a package that you ship with your plugin (I have no idea how to do that or if it's possible) or just deliver a python installer and fetch the library on plugin install/plugin first run/whenever directly from github which is a convenient way to make sure the library is always up to date.
Because you're not beholden to executing stuff at realtime, you can do almost anything with the BackgroundTask class.
Also maybe change that thread title so that it can be useful to someone else in the future
-
@aaronventure Alright, I will look forward to that later. Btw how do I change the Topic's name ?
-
@ClawFORGE-Studio go to your first post in the thread, click the three dots in the bottom right corner of your post, click Edit and change the title above the message entry field.
-
@aaronventure Alright thanks, I'll change it later on, it doesn't allow me to change it for the moment.
-
@ClawFORGE-Studio said in general:
Btw how do I change the Topic's name ?
I don't think you can at the moment because the forum, there are a couple of forum bugs.
@ClawFORGE-Studio said in general:
I can make a Text-To-Sample into HISE as well.
Not with the model you linked, it's trained on speech, you would need to make your own model.
