GitLab

修改默认编译器为hipcc

See merge request dcutoolkit/deeplearing/torchaudio!1

2.添加了dcu_version和相关dtk信息

Summary: Pull Request resolved: https://github.com/pytorch/audio/pull/2911

Reviewed By: carolineechen

Differential Revision: D41887854

Pulled By: mthrok

fbshipit-source-id: eb91773ec67b4cda2d70733df450956d83742509

Summary:
Pull Request resolved: https://github.com/pytorch/audio/pull/2906

The correct way to create AVFormatContext* for output is to pass an address of an uninitialized *AVFormatContext struct to `avformat_alloc_output_context2` function.

The current code pre-allocates AVFormatContext* with `avformat_alloc_context`, then this allocated object is lost inside of `avformat_alloc_output_context2`.

Reviewed By: xiaohui-zhang

Differential Revision: D41865685

fbshipit-source-id: 9a9dc83b5acfe9b450f191fe716c85ebb5a5d842

Summary:
Pull Request resolved: https://github.com/pytorch/audio/pull/2905

In StreamWriter, if the tensor format is different from the encoding format, then a FilterGraph object is automatically inserted to convert the format.

The FilterGraph object operates on AVFrames. The input AVFrame must be allocated by us, but the output AVFrames is filled by FilterGraph, thus no need to allocate it.

Now the output AVFrame is used as input to encoder regardless of whether FilterGraph was inserted. Thus the output AVFrame has to be manually allocated by us when FilterGraph is not used.

The current code flips this condition and incorrectly allocates AVFrame when FilterGraph is present and does not allocate otherwise.

This commit fix that.

Reviewed By: xiaohui-zhang

Differential Revision: D41866198

fbshipit-source-id: 40799c147dc8166a979ecfb58ed8e502539a6aed

Summary:
address https://github.com/pytorch/audio/issues/2885

In `_init_hubert_pretrain_model ` method which initialize the hubert pretrain models, `kaiming_normal_` should be applied on `ConvLayerBlock` instead of `LayerNorm` layer. This PR fixes it and adds more unit tests.

Pull Request resolved: https://github.com/pytorch/audio/pull/2886

Reviewed By: hwangjeff

Differential Revision: D41713801

Pulled By: nateanl

fbshipit-source-id: ed199baf7504d06bbf2d31c522ae708a75426a2d

Summary: Pull Request resolved: https://github.com/pytorch/audio/pull/2865

Reviewed By: carolineechen

Differential Revision: D41403756

Pulled By: mthrok

fbshipit-source-id: d193caa90e786f08f28e4cc2df4b4fb77aa8f592

Summary:
Makes it specific to which version of otool and install_name_tool we actually prefer since using the one from conda can produce inconsistent results

Fixes https://github.com/pytorch/audio/issues/2806

Signed-off-by: Eli Uriegas <eliuriegas@meta.com>

Pull Request resolved: https://github.com/pytorch/audio/pull/2828

Reviewed By: malfet, mthrok

Differential Revision: D40960633

Pulled By: seemethere

fbshipit-source-id: 5010c06578f1efc4fe314f9a3ff47f18e14ad156

Summary:
StreamWriter assumed that frame rate is always expressed as 1/something, which is a reasonable assumption.

This commit fixes it by properly computing time_base from frame rate.

Address https://github.com/pytorch/audio/issues/2830

Pull Request resolved: https://github.com/pytorch/audio/pull/2831

Reviewed By: carolineechen

Differential Revision: D41036084

Pulled By: mthrok

fbshipit-source-id: 805881d4cb221ab2c002563aefb986e30fb91609

Summary:
Addresses https://github.com/pytorch/audio/issues/2790.

Previously AVPacket objects had duration==0.

`av_interleaved_write_frame` function was inferring the duration of packets by
comparing them against the next ones but It could not infer the duration of
the last packet, as there is no subsequent frame, thus was omitting it from the final data.

This commit fixes it by explicitly setting packet duration = 1 (one frame)
only for video. (audio AVPacket contains multiple samples, so it's different.
To ensure the correctness for audio, the tests were added.)

Pull Request resolved: https://github.com/pytorch/audio/pull/2789

Reviewed By: xiaohui-zhang

Differential Revision: D40627439

Pulled By: mthrok

fbshipit-source-id: 4d0d827bff518c017b115445e03bdf0bf1e68320

Summary:
address https://github.com/pytorch/audio/issues/2847

In mixed precision training, the dtype of `mask_embedding` is **not** converted to fp16 automatically. This PR addresses the issue by changing the dtype of `mask_embedding` to `x` to enable mixed precision training.

Pull Request resolved: https://github.com/pytorch/audio/pull/2854

Reviewed By: carolineechen

Differential Revision: D41343486

Pulled By: nateanl

fbshipit-source-id: 4a5cbb429ff8ba5d3c439a3d5acb5094f66bf705

Summary:
- `_get_fileids_paths` in `LibriLightLimited` dataset was changed dataset in https://github.com/pytorch/audio/issues/2653, the absolute path becomes relative paths. This PR fixes the usage in hubert fine-tuning recipe to get correct audio paths.
- model options should be `hubert_pretrain_large` and `hubert_pretrain_xlarge` instead of `hubert_large` and `hubert_xlarge`.
- The input dimension of CTC linear layer varies depending on the model architecture, update it in lightning module.

cc simpleoier

Pull Request resolved: https://github.com/pytorch/audio/pull/2851

Reviewed By: carolineechen

Differential Revision: D41327998

Pulled By: nateanl

fbshipit-source-id: f92248ee84ec860b4e4dbef880c5794b338e1e2d

Summary:
address https://github.com/pytorch/audio/issues/2845

Pull Request resolved: https://github.com/pytorch/audio/pull/2846

Reviewed By: carolineechen

Differential Revision: D41251624

Pulled By: nateanl

fbshipit-source-id: 1a363d2314d6a452f35c109b9730da64ada5a2fd

Summary:
* Add the new official torchaudio logo to documentation/README.
* Add a page for download logo.

https://output.circle-artifacts.com/output/job/e9eb1292-7c10-4fef-adc3-ad568802aa59/artifacts/0/docs/index.html

<img width="1068" alt="Screen Shot 2022-11-14 at 10 30 27 AM" src="https://user-images.githubusercontent.com/855818/201738349-9e248f15-dce2-4931-9066-aa898a53d6ad.png">

https://output.circle-artifacts.com/output/job/e9eb1292-7c10-4fef-adc3-ad568802aa59/artifacts/0/docs/logo.html

<img width="617" alt="Screen Shot 2022-11-14 at 10 30 47 AM" src="https://user-images.githubusercontent.com/855818/201738420-ad0fda2f-f310-4802-851c-bbdf6c84c045.png">

Pull Request resolved: https://github.com/pytorch/audio/pull/2802

Reviewed By: carolineechen

Differential Revision: D41295277

Pulled By: mthrok

fbshipit-source-id: 6615d00799c9611f875e8485459d800e350b3486

Summary: Pull Request resolved: https://github.com/pytorch/audio/pull/2825

Reviewed By: carolineechen

Differential Revision: D40954522

Pulled By: mthrok

fbshipit-source-id: 433fb856a74a340af4d49e5c65a6270f0b00c835

Summary:
PyTorch logo is included in pytorch doc theme, (and cannot be changed without custom CSS) so no need to have them here.

Pull Request resolved: https://github.com/pytorch/audio/pull/2824

Reviewed By: carolineechen

Differential Revision: D40954564

Pulled By: mthrok

fbshipit-source-id: 5e9a91fddcc92c141baf1996f721c09c037fb003

Summary:
<img width="756" alt="Screen Shot 2022-11-01 at 3 32 58 PM" src="https://user-images.githubusercontent.com/855818/199173348-f463ae71-438c-4dad-a481-b65522a8e52f.png">

Pull Request resolved: https://github.com/pytorch/audio/pull/2812

Reviewed By: carolineechen

Differential Revision: D40919942

Pulled By: mthrok

fbshipit-source-id: 18e5a709c262fb0b15ada0d303f1d0dee033beb1

Summary:
address https://github.com/pytorch/audio/issues/2780

Pull Request resolved: https://github.com/pytorch/audio/pull/2781

Reviewed By: carolineechen, mthrok

Differential Revision: D40556794

Pulled By: nateanl

fbshipit-source-id: b24912489d41e5663b4b4dcfb8be743fb962097e

Summary:
add ability to load only improvised or only scripted utterances.

Pull Request resolved: https://github.com/pytorch/audio/pull/2778

Reviewed By: nateanl

Differential Revision: D40511865

Pulled By: carolineechen

fbshipit-source-id: e1fe3908ac2aa306ad30c242ddd25762b2268539

Summary:
previous download link for v0.02 did not download the entire dataset, but only the training dataset, resulting in issues when trying to access the testing or validation data.

Pull Request resolved: https://github.com/pytorch/audio/pull/2777

Reviewed By: nateanl

Differential Revision: D40480605

Pulled By: carolineechen

fbshipit-source-id: a594506b4ccfb548a7d5043b716c58463480c103

Summary:
The file structure of VoxCeleb1 is as follows:
```
root/
└── wav/
    └── speaker_id folders
```
Users who use [Kaldi](https://github.com/kaldi-asr/kaldi/blob/f6f4ccaf213f0fe8b26e633a7dc0c802150626a0/egs/voxceleb/v1/local/make_voxceleb1_v2.pl) to get the VoxCeleb1 dataset have "dev" and "test" folders above "wav" folder. However, in the file lists like https://www.robots.ox.ac.uk/~vgg/data/voxceleb/meta/veri_test.txt or https://www.robots.ox.ac.uk/~vgg/data/voxceleb/meta/iden_split.txt there is not such differentiation. It's not necessary to put the extracted files into separate folders.

This PR adds notes in `VoxCeleb1Identification` and `VoxCeleb1Verification` datasets to inform the file structure to users.

Pull Request resolved: https://github.com/pytorch/audio/pull/2776

Reviewed By: carolineechen

Differential Revision: D40483707

Pulled By: nateanl

fbshipit-source-id: ccd1780a72a5b53f0300c2466c3073a293ad7b8d

Summary: Pull Request resolved: https://github.com/pytorch/audio/pull/2775

Reviewed By: carolineechen

Differential Revision: D40481144

Pulled By: nateanl

fbshipit-source-id: 5d0fb2478767704603a3ec28d74160e7892d4d0e

Summary: Pull Request resolved: https://github.com/pytorch/audio/pull/2774

Reviewed By: carolineechen

Differential Revision: D40445274

Pulled By: nateanl

fbshipit-source-id: 6388323a5fa5c548a86829cb3f7cafee5382d18d

Summary:
* Refactor benchmark script
* Rename `time` variable to avoid (potential) conflicting with time module
* Fix `beta` parameter in benchmark (it was not used previously)
* Use `timeit` module for benchmark
* Add plot
* Move the comment on result at the end
* Add link to an explanation of aliasing

https://output.circle-artifacts.com/output/job/20b57d2f-3614-4161-a18e-e0c1a537739c/artifacts/0/docs/tutorials/audio_resampling_tutorial.html

Pull Request resolved: https://github.com/pytorch/audio/pull/2773

Reviewed By: carolineechen

Differential Revision: D40421337

Pulled By: mthrok

fbshipit-source-id: b402f84d4517695daeca75fb84ad876ef9354b3a

Summary:
In StreamWriter basic usage tutorial, matplotlib is used to generate raster images of waveforms, and the figure used is left unshown in the resulting tutorial with the use of ``sphinx_gallery_defer_figures`` command.

It turned out that this figure is shown in the next code block executed by Sphinx Gallery, and the figure is placed in totally unrelated place. https://pytorch.org/audio/main/tutorials/audio_feature_extractions_tutorial.html

<img width="951" alt="Screen Shot 2022-10-14 at 10 06 58 PM" src="https://user-images.githubusercontent.com/855818/195855124-ecd9be49-5085-4acd-9a93-608d9d1ee9ce.png">

This commit fixes it by closing the figure.

Pull Request resolved: https://github.com/pytorch/audio/pull/2771

Reviewed By: nateanl

Differential Revision: D40382076

Pulled By: mthrok

fbshipit-source-id: 015f2bab8492d3b4fbe70e1174c7776a5aa2679a

Summary:
The separation applies on chunks of audios to avoid OOM. The combination of consecutive chunks is described in the graph:

![image](https://user-images.githubusercontent.com/8653221/195691886-002844e6-4ec5-41de-8910-df8046553998.png)

In the last audio chunk, there is no future chunk to be combined, hence the overlap on the right side doesn't need to be faded.

Pull Request resolved: https://github.com/pytorch/audio/pull/2769

Reviewed By: carolineechen

Differential Revision: D40358382

Pulled By: nateanl

fbshipit-source-id: ec8be895d7a67acb257e2693b64922397163ed5e

Summary:
* Document `__call__` instead of `__init__`
* List CTCHypothesis first as it is used in combination with CTCDecoder
* Fix indentation of score method docstring

Pull Request resolved: https://github.com/pytorch/audio/pull/2766

Reviewed By: carolineechen

Differential Revision: D40349388

Pulled By: mthrok

fbshipit-source-id: 5e512e6c2b29d3533eb62d09b289154ccd1abf4c