Attribute Error: 'NoneType' object has no attribute 'tobytes'

To create a public link, set share=True in launch().
loading weights/Nanami_e600_s2400.pth
gin_channels: 256 self.spk_embed_dim: 109

Traceback (most recent call last):
File "C:\Users\Desktop\Mangio-RVC-v23.7.0_INFER\Mangio-RVC-v23.7.0\infer-web.py", line 316, in vc_single
audio_opt = vc.pipeline(
File "C:\Users\Desktop\Mangio-RVC-v23.7.0_INFER\Mangio-RVC-v23.7.0\vc_infer_pipeline.py", line 564, in pipeline
self.vc(
File "C:\Users\Desktop\Mangio-RVC-v23.7.0_INFER\Mangio-RVC-v23.7.0\vc_infer_pipeline.py", line 450, in vc
(net_g.infer(feats, p_len, pitch, pitchf, sid)[0][0, 0])
RuntimeError: CUDA error: the launch timed out and was terminated
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.

Traceback (most recent call last):
File "C:\Users\Desktop\Mangio-RVC-v23.7.0_INFER\Mangio-RVC-v23.7.0\runtime\lib\site-packages\gradio\routes.py", line 437, in run_predict
output = await app.get_blocks().process_api(
File "C:\Users\Desktop\Mangio-RVC-v23.7.0_INFER\Mangio-RVC-v23.7.0\runtime\lib\site-packages\gradio\blocks.py", line 1349, in process_api
data = self.postprocess_data(fn_index, result["prediction"], state)
File "C:\Users\Desktop\Mangio-RVC-v23.7.0_INFER\Mangio-RVC-v23.7.0\runtime\lib\site-packages\gradio\blocks.py", line 1283, in postprocess_data
prediction_value = block.postprocess(prediction_value)
File "C:\Users\Desktop\Mangio-RVC-v23.7.0_INFER\Mangio-RVC-v23.7.0\runtime\lib\site-packages\gradio\components.py", line 2586, in postprocess
file_path = self.audio_to_temp_file(
File "C:\Users\Desktop\Mangio-RVC-v23.7.0_INFER\Mangio-RVC-v23.7.0\runtime\lib\site-packages\gradio\components.py", line 360, in audio_to_temp_file
temp_dir = Path(dir) / self.hash_bytes(data.tobytes())
AttributeError: 'NoneType' object has no attribute 'tobytes'

I need help ;-;)

Start in infer-web.py at vc_single and follow the call into vc_infer_pipeline.py at pipeline and vc. Reproduce the CUDA launch-timeout path, then verify that failed inference is handled before Gradio's audio postprocessing receives None and that the user gets a useful error instead of the tobytes traceback.