alderpaw / audio

Data manipulation and transformation for audio signal processing, powered by PyTorch

torchaudio: an audio library for PyTorch

• Support audio I/O (Load files, Save files)
  • Load the following formats into a torch Tensor
    • mp3, wav, aac, ogg, flac, avr, cdda, cvs/vms,
    • aiff, au, amr, mp2, mp4, ac3, avi, wmv,
    • mpeg, ircam and any other format supported by libsox.
    • Kaldi (ark/scp)
• Dataloaders for common audio datasets (VCTK, YesNo)
• Common audio transforms
  • Scale, PadTrim, DownmixMono, LC2CL, BLC2CBL, MuLawEncoding, MuLawExpanding

Dependencies

• pytorch (nightly version needed for development)
• libsox v14.3.2 or above
• [optional] vesis84/kaldi-io-for-python commit cb46cb1f44318a5d04d4941cf39084c5b021241e or above

Quick install on OSX (Homebrew):

brew install sox

Linux (Ubuntu):

sudo apt-get install sox libsox-dev libsox-fmt-all

Anaconda

conda install -c conda-forge sox

Installation

# Linux
python setup.py install

# OSX
MACOSX_DEPLOYMENT_TARGET=10.9 CC=clang CXX=clang++ python setup.py install

Quick Usage

import torchaudio
sound, sample_rate = torchaudio.load('foo.mp3')
torchaudio.save('foo_save.mp3', sound, sample_rate) # saves tensor to file

API Reference

API Reference is located here: http://pytorch.org/audio/