Diarization with reverb_v2 works with cpu but NOT with cuda

[dlinford]

I'm using the most recent version of Faster-Whisper-XXL.exe. I can't get Diarization with reverb_v2 to work with CUDA. The transcription works great with CUDA but NOT the diarization. Also diarization works with the cpu but not cuda.

Any ideas?

This is the error that I get.

Traceback (most recent call last):
File "D:\whisper-fast-XXL_main_.py", line 2324, in
File "D:\whisper-fast-XXL_main_.py", line 1990, in cli
File "D:\whisper-fast-XXL_main_.py", line 448, in get_diarization
File "faster_whisper\diarization.py", line 71, in run_model
File "faster_whisper\diarization.py", line 67, in _load_model
File "pyannote\audio\core\pipeline.py", line 344, in to
File "pyannote\audio\core\inference.py", line 192, in to
File "lightning_fabric\utilities\device_dtype_mixin.py", line 55, in to
File "torch\nn\modules\module.py", line 1152, in to
File "torch\nn\modules\module.py", line 802, in _apply
File "torch\nn\modules\rnn.py", line 221, in _apply
File "torch\nn\modules\rnn.py", line 158, in _init_flat_weights
File "torch\nn\modules\rnn.py", line 209, in flatten_parameters
RuntimeError: CUDA error: no kernel image is available for execution on the device
CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1.
Compile with TORCH_USE_CUDA_DSA to enable device-side assertions.

[PYI-14008:ERROR] Failed to execute script 'main' due to unhandled exception!

[Purfview]

Probably your GPU is RTX 50xx, read there: #433