Thank you, mei2.
I installed GIT and used the command line to install whisper jav.
It took me a fair amount of search to find the install path in Windows 10.
C:\Users\*****\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.12_qbz5n2kfra8p0\LocalCache\local-packages\Python312\site-packages\whisperjav
For anyone who's a bit of a noob on command line programs, this is what I had to do to get it to 'work'.
I had to install ffmpeg in the Windows Path.
Download the full ffmpeg pack. Unpack. Rename folder ffmpeg. Move filder to C: (main OS location).
Then open cmd as administrator and type:
setx /m PATH "C:\ffmpeg\bin;%PATH%
After that, you can use the GUI.
For me, the processing stops (it doesn't finish downloading the model) after a while and the program freezes for a long time, but it may work for someone else. I get an error message about the model name.
RuntimeError: Model has been downloaded but the SHA256 checksum does not not match. Please retry loading the model.
This is wrong. The model download doesn't finish. I've tried 3 times and it always fails before full download and the program freezes.
I have models downloaded already.
Can I copy them into Whisperjav?
Right now the program tires to download models in a huggingface folder.
I also get references to a CUDA file that needs to be added to a path.
I can't use the CMD commands shown on the GitHub page.
"'whisperjav' is not recognized as an internal or external command, operable program or batch file."
There is no file called whisperjav anywhere that I can find. Only whisperjav_gui.py and since I can't make that work...
Help?
I am currently using Faster Whisper XXL and have copied the command line examples I've found.
It works fairly well.
I installed GIT and used the command line to install whisper jav.
It took me a fair amount of search to find the install path in Windows 10.
C:\Users\*****\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.12_qbz5n2kfra8p0\LocalCache\local-packages\Python312\site-packages\whisperjav
For anyone who's a bit of a noob on command line programs, this is what I had to do to get it to 'work'.
I had to install ffmpeg in the Windows Path.
Download the full ffmpeg pack. Unpack. Rename folder ffmpeg. Move filder to C: (main OS location).
Then open cmd as administrator and type:
setx /m PATH "C:\ffmpeg\bin;%PATH%
After that, you can use the GUI.
For me, the processing stops (it doesn't finish downloading the model) after a while and the program freezes for a long time, but it may work for someone else. I get an error message about the model name.
RuntimeError: Model has been downloaded but the SHA256 checksum does not not match. Please retry loading the model.
This is wrong. The model download doesn't finish. I've tried 3 times and it always fails before full download and the program freezes.
I have models downloaded already.
Can I copy them into Whisperjav?
Right now the program tires to download models in a huggingface folder.
I also get references to a CUDA file that needs to be added to a path.
I can't use the CMD commands shown on the GitHub page.
"'whisperjav' is not recognized as an internal or external command, operable program or batch file."
There is no file called whisperjav anywhere that I can find. Only whisperjav_gui.py and since I can't make that work...
Help?
I am currently using Faster Whisper XXL and have copied the command line examples I've found.
It works fairly well.
Last edited: