GitLab

There are couple of aspects of this PR that overall improves the maintainability of the code base, based on "decoupling" and "separation of concerns".

First, `sox_effects` functionalities can be either available or unavailable. From the viewpoint of `torchaudio` main module, the looser the connection between the `torchaudio` module and `torchaudio.sox_effects`, the more manageable the code base become because you can change the two modules independently. This is mostly accomplished when the definitions of `initialize_sox` and `shutdown_sox` were moved from `torchaudio.__init__` to `torchaudio.sox_effects`, but the initialization of sox effects are still happening in `torchaudio.__init__`. If we move the initialization to `sox_effects` module, the responsibility of sox initialization is moved to `sox_effects` module, along with the required module availability check etc. The main `torchaudio` module can be carefree about how the `sox_effects` module should work.

In addition to that, I found that `initialize_sox` and `shutdown_sox` are confusing because it sound like they are required for `libsox` based I/O. To make it clear, I renamed them to include `sox_effect` in function name.

Also moving functions from the original places are BC breaking itself, therefore, these functions are re-imported in `torchaudio.__init__` and renamed to match the original names. Therefore the PR is not BC breaking.

It is strange that `sox_backend` depends on custom classes defined in `soudfile_backend`. This PR introduces `backend.common` module where these custom classes are defined and have both `sox_backend` and `soundfile_backend` depend on it.

* Make `_generate_wave_table` device and dtype aware.
Signed-off-by: Bhargav Kathivarapu <bhargavkathivarapu31@gmail.com>

* kaldi compliance files cleanup for spec, fbank, mfcc

* kaldi compliance tests removal for spec, fbank, mfcc
Signed-off-by: Bhargav Kathivarapu <bhargavkathivarapu31@gmail.com>

* also deletes duplicated func

For better cohesion and coupling, move sox_effects related functions to sox_effects module.
Then perform initialization in __init__.py.

See the following for cohesion and coupling.
https://en.wikipedia.org/wiki/Coupling_(computer_programming)

* Replace `backed_guard` with `requires_module`

* Remove backend_guard

* Migrate spectrogram

* Update spectrogram in kaldi.py to support device and dtype

* Remove failing tests
Signed-off-by: Bhargav Kathivarapu <bhargavkathivarapu31@gmail.com>

* Add version compatibility matrix

* Fix device/dtype compatibility of Kaldi mfcc

* Migrate Kaldi mfcc test

* Remove failing tests
Signed-off-by: Bhargav Kathivarapu <bhargavkathivarapu31@gmail.com>

* Add bass with biquad

* Update functional.py

Add the normalization coefficients

* Update test_sox_compatibility.py

In test_sox_compatibility.py file, I add two bass tests: one test sets gain = 30, atol = 1e-4, the other sets gain = 40, atol = 1.5e-4. The details can be seen in pytorch#676

* Update torchscript_consistency_impl.py

Add torchscript test

* Add flake8 test
Co-authored-by: Ji Chen <jimchen90@devfair0160.h2.fair>

* Migrate fbank tests

* Update CCI job environment

* Remove invalid test cases
Signed-off-by: Bhargav Kathivarapu <bhargavkathivarapu31@gmail.com>

* Added the popular GTZAN dataset:

* Added the GTZAN class in torchaudio.datasets using the same format as the rest of the datasets.
* Added the appropriate test function in test_datasets.py.
* Added the GTZAN class in the datasets.rst documentation file.

* Addressed review issues in PR #668

* Added dummy noise .wav in `test/assets/`
* Removed transforms of input and output from the dataset
  `__init__` function, as well as the corresponding methods.
* Replaced rendundant `filtered` and `subset` methods from
  class initialization and also changed the corresponding
  assertion message.

* Fixed E303: too many blank lines error

* Added GTZAN to __init__.__all__

* Fixed incorrectly not importing GTZAN

* removed duplicate warning

* lint
Co-authored-by: Vincent QB <vincentqb@users.noreply.github.com>

* Updates test_transform.py to no longer perform floor division using torch.div

This will prevent this test from breaking when torch.div is updating to throw a runtime error when it would have performed floor division.

* flake8
Co-authored-by: Vincent QB <vincentqb@users.noreply.github.com>

* Add flanger to functional
Signed-off-by: Bhargav Kathivarapu <bhargavkathivarapu31@gmail.com>

* Add random seed
Signed-off-by: Bhargav Kathivarapu <bhargavkathivarapu31@gmail.com>

* fix flanger
Signed-off-by: Bhargav Kathivarapu <bhargavkathivarapu31@gmail.com>

* shape

* Change bool arguments to strings
Signed-off-by: Bhargav Kathivarapu <bhargavkathivarapu31@gmail.com>

* Refactor tests
Signed-off-by: Bhargav Kathivarapu <bhargavkathivarapu31@gmail.com>
Co-authored-by: Vincent QB <vincentqb@users.noreply.github.com>

* Add caching to binary build for more reliable build

* Apply same logic to unittest workflow

* Use env var for switching SoX build and default to not

* Update README

* Fix packaging/CI script

* Add info on README.md for #666

Added a simple command what to do when coming
across undefined reference errors when trying to build
from specific linux+anaconda configurations.

* Separate style check from unit test

* TEST flake8 fail

* TEST clang-format fail

* Revert "TEST flake8 fail"

This reverts commit 88d90185cefe294b9a662129a05c2e26a9844fe4.

* Revert "TEST clang-format fail"

This reverts commit 93517d1ccba650a6ccd813268fd73a4fd63b252c.

* Clean up extension build mechanism and extension location

* Add back the switch to depend on external sox

* Remove print

* Fix

* make parameter optional string.

* raise error if incorrect parameter choice.

* test with slaney.

* Add util to generate whitenoise

* Use sinusoid for pitch and revert vad

* Adopt PyTorch's test utility on kaldi compatibility test

 - Adopt PyTorch's test utility on kaldi compatibility test
 - Separate CPU test and GPU test

* Fix device/dtype conversion

1. Docker image used for smoke test is old (No Python 3.8). Updating this to the latest one. build [here](https://app.circleci.com/pipelines/github/pytorch/audio/1763/workflows/56c846a5-acaa-41a7-92f5-46ec66186c61/jobs/44484)
2. Add `latest` tag to the newly built docker so that later we can always use `latest` version and we do not have to update docker image tag manually.

* Add windows binary jobs

* Make wheel package platform-specific and python-version-specific