torchaudio: an audio library for PyTorch

The aim of torchaudio is to apply PyTorch to the audio domain. By supporting PyTorch, torchaudio follows the same philosophy of providing strong GPU acceleration, having a focus on trainable features through the autograd system, and having consistent style (tensor names and dimension names). Therefore, it is primarily a machine learning library and not a general signal processing library. The benefits of PyTorch can be seen in torchaudio through having all the computations be through PyTorch operations which makes it easy to use and feel like a natural extension.

Support audio I/O (Load files, Save files)
- Load a variety of audio formats, such as wav, mp3, ogg, flac, opus, sphere, into a torch Tensor using SoX
- Kaldi (ark/scp)
Dataloaders for common audio datasets
Common audio transforms
- Spectrogram, AmplitudeToDB, MelScale, MelSpectrogram, MFCC, MuLawEncoding, MuLawDecoding, Resample
Compliance interfaces: Run code using PyTorch that align with other libraries
- Kaldi: spectrogram, fbank, mfcc

Dependencies

PyTorch (See below for the compatible versions)
[optional] vesis84/kaldi-io-for-python commit cb46cb1f44318a5d04d4941cf39084c5b021241e or above

The following are the corresponding torchaudio versions and supported Python versions.

`torch`	`torchaudio`	`python`
`master` / `nightly`	`master` / `nightly`	`>=3.6`, `<=3.9`
`1.9.0`	`0.9.0`	`>=3.6`, `<=3.9`
`1.8.0`	`0.8.0`	`>=3.6`, `<=3.9`
`1.7.1`	`0.7.2`	`>=3.6`, `<=3.9`
`1.7.0`	`0.7.0`	`>=3.6`, `<=3.8`
`1.6.0`	`0.6.0`	`>=3.6`, `<=3.8`
`1.5.0`	`0.5.0`	`>=3.5`, `<=3.8`
`1.4.0`	`0.4.0`	`==2.7`, `>=3.5`, `<=3.8`

Installation

Binary Distributions

To install the latest version using anaconda, run:

conda install -c pytorch torchaudio

To install the latest pip wheels, run:

pip install torchaudio -f https://download.pytorch.org/whl/torch_stable.html

(If you do not have torch already installed, this will default to installing torch from PyPI. If you need a different torch configuration, preinstall torch before running this command.)

Nightly build

Note that nightly build is build on PyTorch's nightly build. Therefore, you need to install the latest PyTorch when you use nightly build of torchaudio.

pip

pip install numpy
pip install --pre torchaudio -f https://download.pytorch.org/whl/nightly/torch_nightly.html

conda

conda install -y -c pytorch-nightly torchaudio

From Source

The build process builds libsox and some codecs that torchaudio need to link to. This is achieve by setting the environment variable BUILD_SOX=1. The build process will fetch and build libmad, lame, flac, vorbis, opus, and libsox before building extension. This process requires cmake and pkg-config.

# Linux
BUILD_SOX=1 python setup.py install

# OSX
BUILD_SOX=1 MACOSX_DEPLOYMENT_TARGET=10.9 CC=clang CXX=clang++ python setup.py install

# Windows
# We need to use the MSVC x64 toolset for compilation, with Visual Studio's vcvarsall.bat or directly with vcvars64.bat.
# These batch files are under Visual Studio's installation folder, under 'VC\Auxiliary\Build\'.
# More information available at:
#   https://docs.microsoft.com/en-us/cpp/build/how-to-enable-a-64-bit-visual-cpp-toolset-on-the-command-line?view=msvc-160#use-vcvarsallbat-to-set-a-64-bit-hosted-build-architecture
call "C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\VC\Auxiliary\Build\vcvarsall.bat" x64 && set BUILD_SOX=0 && python setup.py install
# or
call "C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\VC\Auxiliary\Build\vcvars64.bat" && set BUILD_SOX=0 && python setup.py install

This is known to work on linux and unix distributions such as Ubuntu and CentOS 7 and macOS. If you try this on a new system and find a solution to make it work, feel free to share it by opening an issue.

Quick Usage

import torchaudio

waveform, sample_rate = torchaudio.load('foo.wav')  # load tensor from file
torchaudio.save('foo_save.wav', waveform, sample_rate)  # save tensor to file

Backend Dispatch

By default in OSX and Linux, torchaudio uses SoX as a backend to load and save files. The backend can be changed to SoundFile using the following. See SoundFile for installation instructions.

import torchaudio
torchaudio.set_audio_backend("soundfile")  # switch backend

waveform, sample_rate = torchaudio.load('foo.wav')  # load tensor from file, as usual
torchaudio.save('foo_save.wav', waveform, sample_rate)  # save tensor to file, as usual

Note

SoundFile currently does not support mp3.
"soundfile" backend is not supported by TorchScript.

API Reference

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

Conventions

With torchaudio being a machine learning library and built on top of PyTorch, torchaudio is standardized around the following naming conventions. Tensors are assumed to have "channel" as the first dimension and time as the last dimension (when applicable). This makes it consistent with PyTorch's dimensions. For size names, the prefix n_ is used (e.g. "a tensor of size (n_freq, n_mel)") whereas dimension names do not have this prefix (e.g. "a tensor of dimension (channel, time)")

waveform: a tensor of audio samples with dimensions (channel, time)
sample_rate: the rate of audio dimensions (samples per second)
specgram: a tensor of spectrogram with dimensions (channel, freq, time)
mel_specgram: a mel spectrogram with dimensions (channel, mel, time)
hop_length: the number of samples between the starts of consecutive frames
n_fft: the number of Fourier bins
n_mel, n_mfcc: the number of mel and MFCC bins
n_freq: the number of bins in a linear spectrogram
min_freq: the lowest frequency of the lowest band in a spectrogram
max_freq: the highest frequency of the highest band in a spectrogram
win_length: the length of the STFT window
window_fn: for functions that creates windows e.g. torch.hann_window

Transforms expect and return the following dimensions.

Spectrogram: (channel, time) -> (channel, freq, time)
AmplitudeToDB: (channel, freq, time) -> (channel, freq, time)
MelScale: (channel, freq, time) -> (channel, mel, time)
MelSpectrogram: (channel, time) -> (channel, mel, time)
MFCC: (channel, time) -> (channel, mfcc, time)
MuLawEncode: (channel, time) -> (channel, time)
MuLawDecode: (channel, time) -> (channel, time)
Resample: (channel, time) -> (channel, time)
Fade: (channel, time) -> (channel, time)
Vol: (channel, time) -> (channel, time)

Here, and in the documentation, we use an ellipsis "..." as a placeholder for the rest of the dimensions of a tensor, e.g. optional batching and channel dimensions.

Contributing Guidelines

Please refer to CONTRIBUTING.md

Disclaimer on Datasets

This is a utility library that downloads and prepares public datasets. We do not host or distribute these datasets, vouch for their quality or fairness, or claim that you have license to use the dataset. It is your responsibility to determine whether you have permission to use the dataset under the dataset's license.

If you're a dataset owner and wish to update any part of it (description, citation, etc.), or do not want your dataset to be included in this library, please get in touch through a GitHub issue. Thanks for your contribution to the ML community!

Luckyseven1122/audio