Colab Notebooks

DDSP (Differentiable Digital Signal Processing) enables new interactions with raw audio. This notebook example allows you to convert an input from your microphone into an instrument like a violin or a flute.

This Colab notebook allows you to upload your own audio and transcribe it with one of a few different Transformer-based models, including the MT3 model for multi-instrument audio.

Play with Music Transformer and control the model in a few different ways, or just generate new performances from scratch.

GANSynth generates high-fidelity audio with GANs. Here we generate audio from MIDI while interpolating in timbre space.

MusicVAE learns a latent space of musical sequences. Here we apply the MusicVAE framework to single measures of multi-instrument General MIDI, a symbolic music representation that uses a standard set of 128 instrument sounds.

Performance RNN is an LSTM-based recurrent neural network designed to model polyphonic music with expressive timing and dynamics. This notebook shows you how to generate new performed compositions from a trained model.

This Colab notebook has everything you need to upload your own sounds and use NSynth models to reconstruct and interpolate between them on a GPU at no cost.

MusicVAE learns a latent space of musical scores. This Colab notebook provides functionality for you to randomly sample from the prior distribution and interpolate between existing sequences for several pre-trained MusicVAE models.

Onsets and Frames is an automatic piano music transcription model. This Colab notebook demonstrates running the model on user-supplied recordings.

Code for running experiments related to Latent Constraints: Conditional Generation from Unconditional Generative Models.