pytorch audio issues - Githubissues

pytorch / audio

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

https://pytorch.org/audio

BSD 2-Clause "Simplified" License

2.55k stars 657 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

`kaldi.fbank` does not work with non-contiguous input when `snip_edges=False`

#3856 gau-nernst opened 4 days ago
0
fix: fixing batching in MMS_FA

#3855 ex3ndr opened 1 week ago
2
test

#3854 atalman opened 2 weeks ago
1
Migrate towards linux_lob_v2.yml

#3853 atalman closed 2 weeks ago
4
Can anyone provide a real-time pretrain model for Visual Speech Recognition?

#3852 bernie-122 opened 2 weeks ago
0
[ffmpeg] Print source of StreamReader with error message when decode failed.

#3851 vanviethieuanh opened 1 month ago
3
Fix ffmpeg dlls not being recognised during buildtime

#3850 alinpahontu2912 opened 1 month ago
1
Make some buffers non-persistent (e.g. window, fbank...)

#3849 gau-nernst opened 1 month ago
1
c10::nullopt -> std::nullopt

#3848 r-barnes closed 1 month ago
2
Release 2.5.1 update version

#3847 atalman closed 1 month ago
1
[CODEMOD][pytorch] replace uses of np.ndarray with npt.NDArray (#3845)

#3846 igorsugak closed 1 month ago
2
[Codemod][PSS] Upgrade fbcode/pytorch to Python Scientific Stack 2

#3845 igorsugak opened 1 month ago
3
Docs for 2.5.0 release

#3844 NicolasHug closed 1 month ago
1
[AMD] Hipify torchaudio_decoder

#3843 xw285cornell closed 1 month ago
2
Fix typo in build.ffmpeg.rst

#3842 definitelyuncertain opened 1 month ago
3
[RFC] Support non-GPU hardware-based video decoding and encoding

#3841 cdzhan opened 1 month ago
2
[AMD] hipify torchaudio

#3840 xw285cornell closed 1 month ago
3
[chore] Improve warning logs

#3839 theBeginner86 opened 1 month ago
1
How to train a real-time av-asr pretrain model

#3838 Zhaninh opened 1 month ago
0
Add missing @skipIfNoFFmpeg for TestFileObject

#3837 Tobias-Fischer opened 1 month ago
1
Enable xpu windows CD (#3833)

#3836 chuanqi129 closed 1 week ago
2
Not building CUDA 12.6

#3835 johnnynunez opened 2 months ago
1
Ability to build manylinux2014 compliant wheels for other archs (ppc64le)

#3834 mgiessing opened 2 months ago
0
Enable xpu windows CD

#3833 chuanqi129 closed 2 months ago
1
[Release only] release 2.5 changes

#3832 kit1980 closed 2 months ago
1
feat: reduce computations in backprop of `lfilter`

#3831 yoyolicoris closed 2 months ago
3
Bump actions/download-artifact from 3 to 4.1.7 in /.github/workflows

#3830 dependabot[bot] opened 2 months ago
1
Update version.txt to 2.5.0

#3829 atalman closed 3 months ago
1
Ability to provide initial phase to Griffin-Lim

#3828 aaron-dees opened 3 months ago
0
Prebuilt binaries of torch.audio for aarch64 cuda

#3827 chulkilee opened 3 months ago
1
Adopt aligner from "Huang et al., Less Peaky and More Accurate CTC Forced Alignment by Label Priors"

#3826 dmitry-mli opened 3 months ago
4
torchaudio.transforms.Resample causes Float Point Exception

#3825 zhc7 opened 3 months ago
0
The seek functionality of StreamReader on the video stream does not return the correct frame if the start_time_stamp of the video stream is nonzero.

#3824 w238liu opened 3 months ago
0
StreamWriter doesn't correctly write audio chunks

#3823 arch-user-france1 opened 4 months ago
1
transforms.MFCC results in NaN values on Jetson Orin Nano

#3822 frmser opened 4 months ago
0
Loading Opus files from MLS dataset fails because of file metadata

#3821 niemiaszek opened 4 months ago
0
Replace runners prefix amz2023.

#3820 jeanschmidt opened 4 months ago
1
Replace runners prefix amz2023.

#3819 jeanschmidt opened 4 months ago
1
Replace runners prefix amz2023.

#3818 jeanschmidt opened 4 months ago
1
Docs for torchaudio 2.4

#3817 NicolasHug closed 4 months ago
1
Division by zero in loudness calculation

#3816 DanTremonti opened 4 months ago
0
Division by zero in loudness calculation

#3815 dhanvanth-pk-13760 closed 4 months ago
0
Add xpu linux wheel build into torchaudio build matrix

#3814 chuanqi129 closed 4 months ago
1
Video reading: torchaudio.io.StreamReader seek method returns the first frame, regardless of the input start_timestep (on version 0.13.1)

#3813 StolikTomer opened 4 months ago
0
Use std::optional types

#3812 cyyever opened 4 months ago
1
Fix CUDA 12.5 build

#3811 HennerM opened 4 months ago
3
Loading failure errors should indicate what was being loaded when error occured

#3810 pokepress opened 4 months ago
0
StreamReader.add_basic_video_stream drops last frame if `frame_rate` is specified

#3809 tyler-rt opened 4 months ago
0
Differentiable filtering using a cascade of second order IIR filters

#3808 SuperKogito opened 4 months ago
0
`torchaudio.functional.lfilter` returns `nan` when processing sub-array but not for the whole input array.

#3807 SuperKogito closed 4 months ago
4