torchaudio 2.7.0__cp312-cp312-win_amd64.whl → 2.8.0__cp312-cp312-win_amd64.whl
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Potentially problematic release.
This version of torchaudio might be problematic. Click here for more details.
- torchaudio/__init__.py +16 -5
- torchaudio/_backend/sox.py +2 -2
- torchaudio/_backend/utils.py +33 -0
- torchaudio/_internal/module_utils.py +59 -10
- torchaudio/_torchcodec.py +352 -0
- torchaudio/backend/no_backend.py +2 -2
- torchaudio/backend/soundfile_backend.py +2 -2
- torchaudio/backend/sox_io_backend.py +2 -2
- torchaudio/functional/__init__.py +6 -1
- torchaudio/functional/functional.py +7 -3
- torchaudio/io/__init__.py +10 -3
- torchaudio/kaldi_io.py +6 -0
- torchaudio/lib/_torchaudio.pyd +0 -0
- torchaudio/lib/libtorchaudio.pyd +0 -0
- torchaudio/models/decoder/__init__.py +7 -1
- torchaudio/pipelines/_tts/utils.py +3 -1
- torchaudio/prototype/datasets/musan.py +2 -1
- torchaudio/prototype/functional/_dsp.py +8 -0
- torchaudio/prototype/functional/_rir.py +3 -0
- torchaudio/prototype/functional/functional.py +3 -0
- torchaudio/prototype/models/__init__.py +4 -1
- torchaudio/prototype/models/_conformer_wav2vec2.py +7 -0
- torchaudio/prototype/models/_emformer_hubert.py +4 -0
- torchaudio/prototype/models/conv_emformer.py +4 -0
- torchaudio/prototype/models/hifi_gan.py +6 -0
- torchaudio/prototype/models/rnnt.py +6 -0
- torchaudio/prototype/models/rnnt_decoder.py +3 -0
- torchaudio/prototype/pipelines/__init__.py +11 -2
- torchaudio/prototype/pipelines/_vggish/__init__.py +5 -1
- torchaudio/prototype/pipelines/_vggish/_vggish_impl.py +4 -1
- torchaudio/prototype/pipelines/_vggish/_vggish_pipeline.py +3 -2
- torchaudio/prototype/pipelines/hifigan_pipeline.py +5 -0
- torchaudio/prototype/transforms/_transforms.py +6 -1
- torchaudio/sox_effects/sox_effects.py +4 -1
- torchaudio/transforms/__init__.py +3 -1
- torchaudio/transforms/_transforms.py +3 -2
- torchaudio/utils/download.py +2 -0
- torchaudio/utils/sox_utils.py +19 -0
- torchaudio/version.py +2 -2
- {torchaudio-2.7.0.dist-info → torchaudio-2.8.0.dist-info}/METADATA +136 -124
- {torchaudio-2.7.0.dist-info → torchaudio-2.8.0.dist-info}/RECORD +145 -144
- {torchaudio-2.7.0.dist-info → torchaudio-2.8.0.dist-info}/WHEEL +1 -1
- torio/io/_streaming_media_decoder.py +0 -1
- torio/lib/_torio_ffmpeg4.pyd +0 -0
- torio/lib/_torio_ffmpeg5.pyd +0 -0
- torio/lib/_torio_ffmpeg6.pyd +0 -0
- torio/lib/libtorio_ffmpeg4.pyd +0 -0
- torio/lib/libtorio_ffmpeg5.pyd +0 -0
- torio/lib/libtorio_ffmpeg6.pyd +0 -0
- torio/utils/ffmpeg_utils.py +28 -0
- {torchaudio-2.7.0.dist-info → torchaudio-2.8.0.dist-info/licenses}/LICENSE +0 -0
- {torchaudio-2.7.0.dist-info → torchaudio-2.8.0.dist-info}/top_level.txt +0 -0
torchaudio/utils/download.py
CHANGED
torchaudio/utils/sox_utils.py
CHANGED
|
@@ -1,5 +1,15 @@
|
|
|
1
1
|
"""Module to change the configuration of libsox, which is used by I/O functions like
|
|
2
2
|
:py:mod:`~torchaudio.backend.sox_io_backend` and :py:mod:`~torchaudio.sox_effects`.
|
|
3
|
+
|
|
4
|
+
.. warning::
|
|
5
|
+
Starting with version 2.8, we are refactoring TorchAudio to transition it
|
|
6
|
+
into a maintenance phase. As a result:
|
|
7
|
+
|
|
8
|
+
- Some APIs are deprecated in 2.8 and will be removed in 2.9.
|
|
9
|
+
- The decoding and encoding capabilities of PyTorch for both audio and video
|
|
10
|
+
are being consolidated into TorchCodec.
|
|
11
|
+
|
|
12
|
+
Please see https://github.com/pytorch/audio/issues/3902 for more information.
|
|
3
13
|
"""
|
|
4
14
|
|
|
5
15
|
from typing import Dict, List
|
|
@@ -8,7 +18,9 @@ import torchaudio
|
|
|
8
18
|
|
|
9
19
|
sox_ext = torchaudio._extension.lazy_import_sox_ext()
|
|
10
20
|
|
|
21
|
+
from torchaudio._internal.module_utils import dropping_support
|
|
11
22
|
|
|
23
|
+
@dropping_support
|
|
12
24
|
def set_seed(seed: int):
|
|
13
25
|
"""Set libsox's PRNG
|
|
14
26
|
|
|
@@ -21,6 +33,7 @@ def set_seed(seed: int):
|
|
|
21
33
|
sox_ext.set_seed(seed)
|
|
22
34
|
|
|
23
35
|
|
|
36
|
+
@dropping_support
|
|
24
37
|
def set_verbosity(verbosity: int):
|
|
25
38
|
"""Set libsox's verbosity
|
|
26
39
|
|
|
@@ -38,6 +51,7 @@ def set_verbosity(verbosity: int):
|
|
|
38
51
|
sox_ext.set_verbosity(verbosity)
|
|
39
52
|
|
|
40
53
|
|
|
54
|
+
@dropping_support
|
|
41
55
|
def set_buffer_size(buffer_size: int):
|
|
42
56
|
"""Set buffer size for sox effect chain
|
|
43
57
|
|
|
@@ -50,6 +64,7 @@ def set_buffer_size(buffer_size: int):
|
|
|
50
64
|
sox_ext.set_buffer_size(buffer_size)
|
|
51
65
|
|
|
52
66
|
|
|
67
|
+
@dropping_support
|
|
53
68
|
def set_use_threads(use_threads: bool):
|
|
54
69
|
"""Set multithread option for sox effect chain
|
|
55
70
|
|
|
@@ -63,6 +78,7 @@ def set_use_threads(use_threads: bool):
|
|
|
63
78
|
sox_ext.set_use_threads(use_threads)
|
|
64
79
|
|
|
65
80
|
|
|
81
|
+
@dropping_support
|
|
66
82
|
def list_effects() -> Dict[str, str]:
|
|
67
83
|
"""List the available sox effect names
|
|
68
84
|
|
|
@@ -72,6 +88,7 @@ def list_effects() -> Dict[str, str]:
|
|
|
72
88
|
return dict(sox_ext.list_effects())
|
|
73
89
|
|
|
74
90
|
|
|
91
|
+
@dropping_support
|
|
75
92
|
def list_read_formats() -> List[str]:
|
|
76
93
|
"""List the supported audio formats for read
|
|
77
94
|
|
|
@@ -81,6 +98,7 @@ def list_read_formats() -> List[str]:
|
|
|
81
98
|
return sox_ext.list_read_formats()
|
|
82
99
|
|
|
83
100
|
|
|
101
|
+
@dropping_support
|
|
84
102
|
def list_write_formats() -> List[str]:
|
|
85
103
|
"""List the supported audio formats for write
|
|
86
104
|
|
|
@@ -90,6 +108,7 @@ def list_write_formats() -> List[str]:
|
|
|
90
108
|
return sox_ext.list_write_formats()
|
|
91
109
|
|
|
92
110
|
|
|
111
|
+
@dropping_support
|
|
93
112
|
def get_buffer_size() -> int:
|
|
94
113
|
"""Get buffer size for sox effect chain
|
|
95
114
|
|
torchaudio/version.py
CHANGED
|
@@ -1,2 +1,2 @@
|
|
|
1
|
-
__version__ = '2.
|
|
2
|
-
git_version = '
|
|
1
|
+
__version__ = '2.8.0+cpu'
|
|
2
|
+
git_version = '6e1c7fe9ff6d82b8665d0a46d859d3357d2ebaaa'
|
|
@@ -1,124 +1,136 @@
|
|
|
1
|
-
Metadata-Version: 2.
|
|
2
|
-
Name: torchaudio
|
|
3
|
-
Version: 2.
|
|
4
|
-
Summary: An audio package for PyTorch
|
|
5
|
-
Home-page: https://github.com/pytorch/audio
|
|
6
|
-
Author: Soumith Chintala, David Pollack, Sean Naren, Peter Goldsborough, Moto Hira, Caroline Chen, Jeff Hwang, Zhaoheng Ni, Xiaohui Zhang
|
|
7
|
-
Author-email: soumith@pytorch.org
|
|
8
|
-
Maintainer: Moto Hira, Caroline Chen, Jeff Hwang, Zhaoheng Ni, Xiaohui Zhang
|
|
9
|
-
Maintainer-email: moto@meta.com
|
|
10
|
-
Classifier: Environment :: Plugins
|
|
11
|
-
Classifier: Intended Audience :: Developers
|
|
12
|
-
Classifier: Intended Audience :: Science/Research
|
|
13
|
-
Classifier: License :: OSI Approved :: BSD License
|
|
14
|
-
Classifier: Operating System :: MacOS :: MacOS X
|
|
15
|
-
Classifier: Operating System :: Microsoft :: Windows
|
|
16
|
-
Classifier: Operating System :: POSIX
|
|
17
|
-
Classifier: Programming Language :: C++
|
|
18
|
-
Classifier: Programming Language :: Python :: 3.9
|
|
19
|
-
Classifier: Programming Language :: Python :: 3.10
|
|
20
|
-
Classifier: Programming Language :: Python :: 3.11
|
|
21
|
-
Classifier: Programming Language :: Python :: 3.12
|
|
22
|
-
Classifier: Programming Language :: Python :: 3.13
|
|
23
|
-
Classifier: Programming Language :: Python :: Implementation :: CPython
|
|
24
|
-
Classifier: Topic :: Multimedia :: Sound/Audio
|
|
25
|
-
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
|
|
26
|
-
Description-Content-Type: text/markdown
|
|
27
|
-
License-File: LICENSE
|
|
28
|
-
Requires-Dist: torch==2.
|
|
29
|
-
Dynamic: author
|
|
30
|
-
Dynamic: author-email
|
|
31
|
-
Dynamic: classifier
|
|
32
|
-
Dynamic: description
|
|
33
|
-
Dynamic: description-content-type
|
|
34
|
-
Dynamic: home-page
|
|
35
|
-
Dynamic:
|
|
36
|
-
Dynamic: maintainer
|
|
37
|
-
Dynamic:
|
|
38
|
-
Dynamic:
|
|
39
|
-
|
|
40
|
-
|
|
41
|
-
|
|
42
|
-
|
|
43
|
-
|
|
44
|
-
[](https://pytorch.org/audio/main/)
|
|
45
|
+
[](https://anaconda.org/pytorch/torchaudio)
|
|
46
|
+
[](https://anaconda.org/pytorch/torchaudio)
|
|
47
|
+
|
|
48
|
+
|
|
49
|
+
|
|
50
|
+
> [!NOTE]
|
|
51
|
+
> **We are in the process of refactoring TorchAudio and transitioning it into a
|
|
52
|
+
> maintenance phase. This process will include removing some user-facing
|
|
53
|
+
> features: those features are deprecated from TorchAudio 2.8 and will be removed in 2.9.
|
|
54
|
+
> Our main goals are to reduce redundancies with the rest of the
|
|
55
|
+
> PyTorch ecosystem, make it easier to maintain, and create a version of
|
|
56
|
+
> TorchAudio that is more tightly scoped to its strengths: processing audio
|
|
57
|
+
> data for ML. Please see
|
|
58
|
+
> [our community message](https://github.com/pytorch/audio/issues/3902)
|
|
59
|
+
> for more details.**
|
|
60
|
+
|
|
61
|
+
The aim of torchaudio is to apply [PyTorch](https://github.com/pytorch/pytorch) to
|
|
62
|
+
the audio domain. By supporting PyTorch, torchaudio follows the same philosophy
|
|
63
|
+
of providing strong GPU acceleration, having a focus on trainable features through
|
|
64
|
+
the autograd system, and having consistent style (tensor names and dimension names).
|
|
65
|
+
Therefore, it is primarily a machine learning library and not a general signal
|
|
66
|
+
processing library. The benefits of PyTorch can be seen in torchaudio through
|
|
67
|
+
having all the computations be through PyTorch operations which makes it easy
|
|
68
|
+
to use and feel like a natural extension.
|
|
69
|
+
|
|
70
|
+
- [Support audio I/O (Load files, Save files)](http://pytorch.org/audio/main/)
|
|
71
|
+
- Load a variety of audio formats, such as `wav`, `mp3`, `ogg`, `flac`, `opus`, `sphere`, into a torch Tensor using SoX
|
|
72
|
+
- [Kaldi (ark/scp)](http://pytorch.org/audio/main/kaldi_io.html)
|
|
73
|
+
- [Dataloaders for common audio datasets](http://pytorch.org/audio/main/datasets.html)
|
|
74
|
+
- Audio and speech processing functions
|
|
75
|
+
- [forced_align](https://pytorch.org/audio/main/generated/torchaudio.functional.forced_align.html)
|
|
76
|
+
- Common audio transforms
|
|
77
|
+
- [Spectrogram, AmplitudeToDB, MelScale, MelSpectrogram, MFCC, MuLawEncoding, MuLawDecoding, Resample](http://pytorch.org/audio/main/transforms.html)
|
|
78
|
+
- Compliance interfaces: Run code using PyTorch that align with other libraries
|
|
79
|
+
- [Kaldi: spectrogram, fbank, mfcc](https://pytorch.org/audio/main/compliance.kaldi.html)
|
|
80
|
+
|
|
81
|
+
Installation
|
|
82
|
+
------------
|
|
83
|
+
|
|
84
|
+
Please refer to https://pytorch.org/audio/main/installation.html for installation and build process of TorchAudio.
|
|
85
|
+
|
|
86
|
+
|
|
87
|
+
API Reference
|
|
88
|
+
-------------
|
|
89
|
+
|
|
90
|
+
API Reference is located here: http://pytorch.org/audio/main/
|
|
91
|
+
|
|
92
|
+
Contributing Guidelines
|
|
93
|
+
-----------------------
|
|
94
|
+
|
|
95
|
+
Please refer to [CONTRIBUTING.md](./CONTRIBUTING.md)
|
|
96
|
+
|
|
97
|
+
Citation
|
|
98
|
+
--------
|
|
99
|
+
|
|
100
|
+
If you find this package useful, please cite as:
|
|
101
|
+
|
|
102
|
+
```bibtex
|
|
103
|
+
@article{yang2021torchaudio,
|
|
104
|
+
title={TorchAudio: Building Blocks for Audio and Speech Processing},
|
|
105
|
+
author={Yao-Yuan Yang and Moto Hira and Zhaoheng Ni and Anjali Chourdia and Artyom Astafurov and Caroline Chen and Ching-Feng Yeh and Christian Puhrsch and David Pollack and Dmitriy Genzel and Donny Greenberg and Edward Z. Yang and Jason Lian and Jay Mahadeokar and Jeff Hwang and Ji Chen and Peter Goldsborough and Prabhat Roy and Sean Narenthiran and Shinji Watanabe and Soumith Chintala and Vincent Quenneville-Bélair and Yangyang Shi},
|
|
106
|
+
journal={arXiv preprint arXiv:2110.15018},
|
|
107
|
+
year={2021}
|
|
108
|
+
}
|
|
109
|
+
```
|
|
110
|
+
|
|
111
|
+
```bibtex
|
|
112
|
+
@misc{hwang2023torchaudio,
|
|
113
|
+
title={TorchAudio 2.1: Advancing speech recognition, self-supervised learning, and audio processing components for PyTorch},
|
|
114
|
+
author={Jeff Hwang and Moto Hira and Caroline Chen and Xiaohui Zhang and Zhaoheng Ni and Guangzhi Sun and Pingchuan Ma and Ruizhe Huang and Vineel Pratap and Yuekai Zhang and Anurag Kumar and Chin-Yun Yu and Chuang Zhu and Chunxi Liu and Jacob Kahn and Mirco Ravanelli and Peng Sun and Shinji Watanabe and Yangyang Shi and Yumeng Tao and Robin Scheibler and Samuele Cornell and Sean Kim and Stavros Petridis},
|
|
115
|
+
year={2023},
|
|
116
|
+
eprint={2310.17864},
|
|
117
|
+
archivePrefix={arXiv},
|
|
118
|
+
primaryClass={eess.AS}
|
|
119
|
+
}
|
|
120
|
+
```
|
|
121
|
+
|
|
122
|
+
Disclaimer on Datasets
|
|
123
|
+
----------------------
|
|
124
|
+
|
|
125
|
+
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.
|
|
126
|
+
|
|
127
|
+
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!
|
|
128
|
+
|
|
129
|
+
Pre-trained Model License
|
|
130
|
+
-------------------------
|
|
131
|
+
|
|
132
|
+
The pre-trained models provided in this library may have their own licenses or terms and conditions derived from the dataset used for training. It is your responsibility to determine whether you have permission to use the models for your use case.
|
|
133
|
+
|
|
134
|
+
For instance, SquimSubjective model is released under the Creative Commons Attribution Non Commercial 4.0 International (CC-BY-NC 4.0) license. See [the link](https://zenodo.org/record/4660670#.ZBtWPOxuerN) for additional details.
|
|
135
|
+
|
|
136
|
+
Other pre-trained models that have different license are noted in documentation. Please checkout the [documentation page](https://pytorch.org/audio/main/).
|