Closed mravanelli closed 10 months ago
Hi @mravanelli
Thanks for the report. This is related to backend migration. It seems that FFmpeg does not handle certain number of channels well, because FFmpeg considers channel_layout as first citizen and the number of channels as derived property of channel layout, and FFmpeg thinks that there is no good channel_layout defined for 17 channels.
Can you provide backend="sox"
to these calls? It will fallback to the same old sox backend, and it should work.
Yes, with sox it works. It is something related to FFmpeg. It is not a big issue though. I think it would be great to raise an error easier to interpret when people are using the FFmpeg backend with more than 16 channels.
On Tue, Oct 17, 2023 at 2:21 PM moto @.***> wrote:
Hi @mravanelli https://github.com/mravanelli
Thanks for the report. This is related to backend migration. It seems that FFmpeg does not handle certain number of channels well, because FFmpeg considers channel_layout as first citizen and the number of channels as derived property of channel layout, and FFmpeg thinks that there is no good channel_layout defined for 17 channels.
Can you provide backend="sox" to these calls? It will fallback to the same old sox backend, and it should work.
— Reply to this email directly, view it on GitHub https://github.com/pytorch/audio/issues/3659#issuecomment-1766936396, or unsubscribe https://github.com/notifications/unsubscribe-auth/AEA2ZVUIJCQ5TOZV46S7T73X73EC3AVCNFSM6AAAAAA6BIICLOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTONRWHEZTMMZZGY . You are receiving this because you were mentioned.Message ID: @.***>
🐛 Describe the bug
Hi, I noticed this weird behavior and I would like to report:
Versions
Versions of relevant libraries: [pip3] flake8==3.7.9 [pip3] mypy-extensions==0.4.3 [pip3] numpy==1.21.5 [pip3] numpydoc==1.4.0 [pip3] torch==2.1.0 [pip3] torchaudio==2.1.0 [pip3] torchinfo==1.8.0 [pip3] torchvision==0.16.0 [pip3] triton==2.1.0 [conda] blas 1.0 mkl
[conda] mkl 2021.4.0 h06a4308_640
[conda] mkl-service 2.4.0 py39h7f8727e_0
[conda] mkl_fft 1.3.1 py39hd3c417c_0
[conda] mkl_random 1.2.2 py39h51133e4_0
[conda] numpy 1.21.5 py39h6c91a56_3
[conda] numpy-base 1.21.5 py39ha15fc14_3
[conda] numpydoc 1.4.0 py39h06a4308_0
[conda] torch 2.1.0 pypi_0 pypi [conda] torchaudio 2.1.0 pypi_0 pypi [conda] torchinfo 1.8.0 pypi_0 pypi [conda] torchvision 0.16.0 pypi_0 pypi [conda] triton 2.1.0 pypi_0 pypi