We’re pleased to present Neutone Morpho, a realtime tone morphing plugin. Our cutting-edge machine learning technology can transform any sound into something new and inspiring.
Neutone Morpho directly processes audio, capturing even the subtlest details from your input.
Unleash your creativity with Neutone Morpho!
What is AI Tone Morphing?
With our pre-trained AI models you can transform any incoming audio into the characteristics, or “style”, of the sounds that the model is based on. In real-time! Sometimes this leads to surprising outcomes. Listen to the examples below.
input
input
output
click on one of the inputs first
output
The Morpho Model
The Neutone Morpho Algorithm utilizes an autoencoder AI algorithm consisting of an encoder and decoder to transform audio. The encoder analyzes the incoming audio, outputting information about the pitch, loudness and other more subtle features of the incoming audio (this is sometimes referred to as the latent variable). The model is trained on specific types of sounds, such as violin performances, and can create a tone transfer effect by interpreting different input sounds in the context of the trained data.
Morpho Model
At the core of Neutone Morpho are the Morpho AI models, where the magic happens. You can interact with a loaded Morpho model in two modes to influence the tone-morphing process.
In Macro Mode, you have four macro knobs that allow you to change the general behaviors of the model. For example, the “Serendipity” knob determines the level of randomness in the transfer, with higher serendipity resulting in more complex sound. Each model has its own unique macros. For instance, a percussion model may have a knob called “decay” that controls the decay of each hit.
The Morpho AI model crunches incoming audio into 6 or more values, referred to as “latent variables.” In Micro View, you can directly offset and scale those values with Morpho Modifiers to steer the output in certain directions. You may even uncover a hidden correlation between a latent variable and the pitch of the output! (More technical details are available here.)
Pre- & Post Processing
Here is a list of pre/post processing DSP (Digital Signal Processing) features designed to achieve optimal performance of Morpho models. Please note that you can enable or disable these features using the small orange power button at the top right of each section.
Pricing
We're giving you a fully working version for free to test out. There is no time limit, so feel free to play around with it as much as you want. If you enjoy it and want to use more models or try out custom model training with Neutone Cocoon, go ahead and upgrade to the full version.
Free Demo
Full ($99)
Bundled Models
4 Models
5 additional Models Early Supporter: 10 additional Models!
Neutone Morpho v1.0.1
Features and Changes:
- Delay Sync Mode
- The Preprocessing Delay module can now toggle between milliseconds and notes
- When the DAW buffer size is set to 256 and below there is now a warning message to indicate implications on performance
- Tooltips added to Macro knobs
- Windows install location now configurable
- Parameter Automation is now correctly reflected on UI
- Minor bug fixes relating to presets and error handling
Requirements
Mac
Mac OS 12.0+
Intel or Apple Silicon
Plugin Format: VST3/AU
Windows
Windows 10 & 11
Plugin Format: VST3
Installation
Download the installer for your platform
Run the installer
Tested DAWs
Ableton Live 11 & 12
Logic Pro 10
Bitwig Studio 5
Studio One 6
Cubase 13
Reaper 7
We respect the copyrights of artists and musicians, as well as the efforts they put into creating music and sound materials. We do not use any copyrighted materials for model training. Morpho models are trained using publicly available training data that is either in the public domain or licensed under Creative Commons. For more details, please refer to the model card of each specific model.
Read our post about our stance on AI and copyright
A: The behavior of each model is highly dependent on the nature of the data on which it was trained. Models trained on percussive sound perform well with percussive input, while models trained on sustained sound tend to perform well with sustained input.
If you find the output of a model unsatisfactory, please try input sounds that have similar characteristics to the model's training data.
A: With Neutone Morpho we created an instrument that is specialised on tone morphing. The "old" Neutone FX plugin can handle many different types of machine learning architecture.
We plan to keep Neutone FX free and will maintain in the future so that audio researchers have an easy way to put their AI models into the DAW.
A: Your models are saved in your User folder. You can change the location in the plugin settings.
On Mac: ~/Library/Application Support/Neutone/NeutoneMorpho
On Windows: C:\Users\<your username>\AppData\Roaming\Neutone\NeutoneMorpho\
A: Of course
A: Morpho models are trained using publicly available training data that is either in the public domain or licensed under Creative Commons. For more details, refer to the model card of each specific model.
A: Yes, in the near future! We are currently developing a cloud-based model training service called Neutone Cocoon. To train your own model, simply collect and upload your training data to our server. Once the training is complete, the trained model will be available in your model browser.
Neutone Cocoon - Cloud-GPU Model training service (coming soon…)
We are currently developing a cloud-based model training service called Neutone Cocoon. To train your own model, simply collect and upload your training data to our server. Once the training is complete, the trained model will be available in your model browser. The trained model will only be available to you, and you own the model.
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!