Community-based Platform where AI Researchers and Engineers Can Share Deep Learning DSP Models with Creators
We release Neutone β, a plugin that makes real-time DSP (digital sound processing) models using deep learning available on DAWs (digital audio workstations). The general-purpose plugin enables artists to easily integrate AI powered tools into their creative process. AI researchers and engineers can easily share newly developed models with music makers through this platform.
■ The Main Features of Neutone Plugin
Up until this point it has been necessary to learn a programming language such as Python and use dedicated hardware such as a GPU to be able to experiment with AI. Using the VST3 and AU formats, Neutone can be run in real-time on the CPU in DAWs such as Ableton Live, Reaper, Logic Pro and others. A number of timbre transfer models and deep learning powered effects have already been published by Qosmo for this plugin and can be downloaded at runtime.
Below are descriptions of some of the models currently available：
Some of these models are still in the experimental phase but there are ongoing efforts to both improve the quality of existing ones as well as increasing the range of available models. For more information on these models, please check the official Neutone website.
For those who are AI researchers and developers, you can publish your own models on the Neutone platform. By wrapping your models with the NeutoneSDK provided by us, you can easily make them compatible with the Neutone plugin.
■Co-created AI models within the AI music community
Neutone provides a platform for hosting AI models, but we would like to co-create AI models with other researchers, engineers, and artists. If you want to be a part of this emerging community come and join us in the Neutone Discord channel.