Blank error when attempting to prepare dataset for training. #171

New Issue

The13thDrifter · 2023-03-24T20:23:57Z

The13thDrifter commented

2023-03-24 20:23:57 +00:00

Just like it says in the title, whenever I attempt to prepare a dataset for voice training, I get given a completely blank error code- so I have no idea where to begin on fixing it. Here is the Command Prompt log.

Traceback (most recent call last):
File "C:\Users\jenik\Documents\FourChanFork\ai-voice-cloning\venv\lib\site-packages\gradio\routes.py", line 394, in run_predict
output = await app.get_blocks().process_api(
File "C:\Users\jenik\Documents\FourChanFork\ai-voice-cloning\venv\lib\site-packages\gradio\blocks.py", line 1075, in process_api
result = await self.call_function(
File "C:\Users\jenik\Documents\FourChanFork\ai-voice-cloning\venv\lib\site-packages\gradio\blocks.py", line 884, in call_function
prediction = await anyio.to_thread.run_sync(
File "C:\Users\jenik\Documents\FourChanFork\ai-voice-cloning\venv\lib\site-packages\anyio\to_thread.py", line 31, in run_sync
return await get_asynclib().run_sync_in_worker_thread(
File "C:\Users\jenik\Documents\FourChanFork\ai-voice-cloning\venv\lib\site-packages\anyio_backends_asyncio.py", line 937, in run_sync_in_worker_thread
return await future
File "C:\Users\jenik\Documents\FourChanFork\ai-voice-cloning\venv\lib\site-packages\anyio_backends_asyncio.py", line 867, in run
result = context.run(func, *args)
File "C:\Users\jenik\Documents\FourChanFork\ai-voice-cloning\src\webui.py", line 201, in prepare_dataset_proxy
message = transcribe_dataset( voice=voice, language=language, skip_existings=skip_existings, progress=progress )
File "C:\Users\jenik\Documents\FourChanFork\ai-voice-cloning\src\utils.py", line 1270, in transcribe_dataset
files = sorted( get_voices(load_latents=False)[voice] )
KeyError: ''

Just like it says in the title, whenever I attempt to prepare a dataset for voice training, I get given a completely blank error code- so I have no idea where to begin on fixing it. Here is the Command Prompt log. > Traceback (most recent call last): File "C:\Users\jenik\Documents\FourChanFork\ai-voice-cloning\venv\lib\site-packages\gradio\routes.py", line 394, in run_predict output = await app.get_blocks().process_api( File "C:\Users\jenik\Documents\FourChanFork\ai-voice-cloning\venv\lib\site-packages\gradio\blocks.py", line 1075, in process_api result = await self.call_function( File "C:\Users\jenik\Documents\FourChanFork\ai-voice-cloning\venv\lib\site-packages\gradio\blocks.py", line 884, in call_function prediction = await anyio.to_thread.run_sync( File "C:\Users\jenik\Documents\FourChanFork\ai-voice-cloning\venv\lib\site-packages\anyio\to_thread.py", line 31, in run_sync return await get_asynclib().run_sync_in_worker_thread( File "C:\Users\jenik\Documents\FourChanFork\ai-voice-cloning\venv\lib\site-packages\anyio\_backends\_asyncio.py", line 937, in run_sync_in_worker_thread return await future File "C:\Users\jenik\Documents\FourChanFork\ai-voice-cloning\venv\lib\site-packages\anyio\_backends\_asyncio.py", line 867, in run result = context.run(func, *args) File "C:\Users\jenik\Documents\FourChanFork\ai-voice-cloning\src\webui.py", line 201, in prepare_dataset_proxy message = transcribe_dataset( voice=voice, language=language, skip_existings=skip_existings, progress=progress ) File "C:\Users\jenik\Documents\FourChanFork\ai-voice-cloning\src\utils.py", line 1270, in transcribe_dataset files = sorted( get_voices(load_latents=False)[voice] ) KeyError: ''

👍 1

psammites commented

2023-03-24 20:34:18 +00:00

What do you have set as your Whisper Backend?

Edit: Did you run the appropriate setup script for your hardware?

What do you have set as your Whisper Backend? Edit: Did you run the appropriate setup script for your hardware?

The13thDrifter commented

2023-03-24 20:41:48 +00:00

Yes, I ran the appropriate setup script; windows CUDA- as for my Whisper Backend- it's set to openAI

psammites commented

2023-03-24 20:47:54 +00:00

Hmm. What's in the directory for the voice you're attempting to prepare the dataset from? Are the files valid .wav's?

The13thDrifter commented

2023-03-24 20:52:38 +00:00

The directory leads directly to where the .wav files are stored. I'm attempting to train a Cortana voice model from the Halo 1 voice files. The Wavs have all been converted to the correct format.

D:\SteamLibrary\steamapps\common\xVATrainer\resources\app\datasets\halo_cortana\wavs

Note that the 'wavs' at the end is referring to a folder named 'wavs', that has the .wav files inside of it. If it helps, the blank error came after a couple of seconds where it looked to be processing originally. Now, the error is instant.

The directory leads directly to where the .wav files are stored. I'm attempting to train a Cortana voice model from the Halo 1 voice files. The Wavs have all been converted to the correct format. > D:\SteamLibrary\steamapps\common\xVATrainer\resources\app\datasets\halo_cortana\wavs Note that the 'wavs' at the end is referring to a folder named 'wavs', that has the .wav files inside of it. If it helps, the blank error came after a couple of seconds where it looked to be processing originally. Now, the error is instant.

mrq commented

2023-03-24 20:53:13 +00:00

ai-voice-cloning\src\utils.py", line 1270

Well, to start with, you're on an outdated version, so I suggest updating, as that line is nowhere near correct in the upstream version.

> [`ai-voice-cloning\src\utils.py", line 1270`](https://git.ecker.tech/mrq/ai-voice-cloning/src/branch/master/src/utils.py#L1270) Well, to start with, you're on an outdated version, so I suggest updating, as that line is nowhere near correct in the upstream version.

mrq commented

2023-03-24 20:55:23 +00:00

D:\SteamLibrary\steamapps\common\xVATrainer\resources\app\datasets\halo_cortana\wavs

Per the documentation:

Dataset Source: a valid folder under ./voice/, as if you were using it to generate with.

It's not a text field, it's a dropdown of names in the voice folder.

> D:\SteamLibrary\steamapps\common\xVATrainer\resources\app\datasets\halo_cortana\wavs Per [the documentation](https://git.ecker.tech/mrq/ai-voice-cloning/wiki/Training#prepare-dataset): > Dataset Source: a valid folder under ./voice/, as if you were using it to generate with. It's not a text field, it's a dropdown of names in the voice folder.

The13thDrifter commented

2023-03-24 20:57:58 +00:00

Ah shit, I must have skipped over that; sorry, it was way too early when I set this up. I'll move it over and check to see if the error repeats with the correct path

sazandora commented

2023-03-24 21:13:07 +00:00

Please, keep me posted! Getting this on Colab, too. It makes sense, since there's no actual option to refresh the "dataset source" list, there was no way for me to select my voice from the list, so I just entered a path, as well.

psammites commented

2023-03-24 21:21:08 +00:00

It makes sense, since there's no actual option to refresh the "dataset source" list, there was no way for me to select my voice from the list

"Refresh Voice List" on the Generate tab will do it.

> It makes sense, since there's no actual option to refresh the "dataset source" list, there was no way for me to select my voice from the list "Refresh Voice List" on the Generate tab will do it.

The13thDrifter commented

2023-03-24 21:40:36 +00:00

Okay, I'm thinking it was user error. I've just updated the software and changed the .wav file location, and I'm no-longer getting an blank error. I'm instead getting a 'missing whisper.json' error- but I know that the fix for that is located in the wiki. Thanks again for the help.

The13thDrifter closed this issue

2023-03-25 02:25:17 +00:00

Sign in to join this conversation.