Unable to seperate vocal and train models

Hello,
Whenever I try to seperate vocal this message appeared.

Traceback (most recent call last):
File "C:\Users\user\Downloads\RVC1006Nvidia\infer\modules\uvr5\modules.py", line 41, in uvr
paths = [os.path.join(inp_root, name) for name in os.listdir(inp_root)]
NotADirectoryError: [WinError 267] The directory name is invalid: 'C:\Users\user\Downloads\VID 1.mp3'

Similarly, when I try to train a model, step 2A yield
Traceback (most recent call last):
File "C:\Users\user\Downloads\RVC1006Nvidia\infer\modules\train\preprocess.py", line 11, in
sr = int(sys.argv[2])
ValueError: invalid literal for int() with base 10: 'folder\VID'
2024-11-24 21:09:59 | INFO | main |

If I ignore all the error and proceed with the actual training process, it failed of course and upon insepecting the logs, seems like a lot of file location error is going on. " The system cannot find the path specified: 'C:\Users\user\Downloads\RVC1006Nvidia/logs/mi-test/1_16k_wavs'"

Should I put my RVC1006Nvidia foldier else where or I should place the audio file in a speific location for it to work?

Start with infer/modules/uvr5/modules.py and infer/modules/train/preprocess.py, then reproduce the reported Windows path errors using the vocal-separation and step 2A workflows. Determine whether the input paths are expected to be files or directories and whether the reported failures indicate a product bug or an undocumented setup requirement.