A Torch implementation of SampleRNN: An Unconditional End-to-End Neural Audio Generation Model.

Sample output can be heard here. Feel free to submit links to any interesting samples you generate as a pull request.

The following packages are required to run SampleRNN_torch:

• nn
• cunn
• cudnn
• rnn
• optim
• audio
• xlua
• gnuplot

To retrieve and prepare the piano dataset, as used in the reference implementation, run:

cd datasets/piano/
./create_piano_dataset.sh

The violin dataset preparation scripts are located in datasets/violin/.

Custom datasets may be created by using scripts/generate_dataset.lua to slice multiple audio files into segments for training, audio must be placed in datasets/[dataset]/data/.

To start a training session run th train.lua -dataset piano. To view a description of all accepted arguments run th train.lua -help.

To view the progress of training run th generate_plots, the loss and gradient norm curve will be saved in sessions/[session]/plots/.

By default samples are generated at the end of every training epoch but they can also be generated separately using th train.lua -generate_samples with the session parameter to specify the model.

Multiple samples are generated in batch mode for efficiency, however generating a single audio sample is faster with th fast_sample.lua. See -help for a description of the arguments.

A pretrained model of the piano dataset is available here. Download and copy it into your sessions/ directory and then extract it in place.

More models will be uploaded soon.