Introducing Neutone Cocoon
Cloud-GPU Model training service
Neutone Cocoon is a cloud-based model service within the Neutone ecosystem. The process is simple: upload your audio materials, and we'll train a unique Neutone model for you on our GPU server. The trained model will be available for you (and you only) to use on Neutone plugins
Currently, Neutone Cocoon supports the training of Neutone Morpho models.
We firmly believe in the importance of personalized AI models for creative endeavors. Each artist should have their own unique tools for artistic expression. We are delighted to support this trend!
Access to Neutone Cocoon is currently limited to selected artists.
If you're eager to create your own models for Neutone Morpho, join the waitlist and let us know what you would like to train on.
Data Collection
To train a model you need a good dataset
Data amount
More than 1 hour of sound data is ideal The dataset can be a collection of many short snippets or a single long recording
Diversity of Data
Too little diversity results in models that output the same thing over and over. Too much diversity and the RAVE model tends to fail. Compiling recordings of a single instrument should generally be just the right amount of diversity.
Recording quality
The audio should ideally be recorded in a clean environment. All the sounds should be recorded in a similar environment.
Range
The pitch range of the input affects the model’s reactivity. Models tend to break and become unstable when it is fed input that has frequency content that was not in the training data.
Preprocessing
Preprocessing your data before training may be effective in improving model quality.
Gain normalization is necessary if the original data is relatively quiet.
A model trained on quiet sounds can behave erratically when it is fed loud sounds as input. If your data doesn’t cover a wide enough range of pitch, pitch augmentation (creating variations by pitch-shifting the data) may be beneficial.
This is also a good way to increase the amount of data. You can use predefined preprocessing procedures on Neutone Cocoon.
Please abide by copy protection laws and to our AI ethics standards when collecting and using data.
- Although the legalities surrounding generative AI are constantly evolving, please refrain from using recordings of other artists without their consent.
- Please read and abide by our terms and conditions.
- We may refuse your request based on the audio files you submitted