“Our goal with TextToSample is to enable the producer community to experiment with state-of-the-art generative AI models. We want to show what’s currently possible with these models, especially when running them locally on your computer. TextToSample’s purpose is certainly not to replace human musicians, which it’s not capable of doing anyway.
TextToSample uses Meta’s MusicGen model weights, which are licenced under CC BY-NC 4.0. In particular, the weights of the “small” model are used. You can find more information about the licence and intended use here.
All calculations are done locally on your computer. You only need an Internet connection to download the models when you open TextToSample for the first time.
An Apple M1 chip with Neural Engine or a GPU is highly recommended. Otherwise the model will take a long time to generate audio.
- Windows: 10 or higher
- MacOS: 12 or higher
Unfortunately we’re not able to get a VST2 license because it was discontinued and only developers who created VST2 plugins in the past are still able to distribute them. If you really need a VST2 plugin, we recommend using a VST3 to VST2 bridge such as Kushview’s Element. We are not affiliated with Kushview and are not taking any responsibility for their product to work with TextToSample.”
Source : samplab.com
