• https://git.ecker.tech/ aims to provide a place to share my efforts while maintaining true ownership of my code, as I do not trust GitHub.

    XMR: 4B9TQdkAkBFYrbj5ztvTx89e5LpucPeTSPzemCihdDi9EBnx7btn8RDNZTBz2zihWsjMnDkzn5As1LU6gLv3KQy8BLsZ8SG

  • Joined on 2022-10-10
ecker commented on issue ecker/ai-voice-cloning#15 2023-02-20 23:03:32 +00:00
Colab: ModuleNotFoundError: No module named 'tortoise'

Fixed in commit 1fd88afccad45f5579412c9c380eb0e9cf59ee6d. Just finished validating it.

ecker closed issue ecker/ai-voice-cloning#15 2023-02-20 23:03:32 +00:00
Colab: ModuleNotFoundError: No module named 'tortoise'
ecker pushed to master at ecker/ai-voice-cloning 2023-02-20 22:56:53 +00:00
1fd88afcca updated notebook for newer setup structure, added formatting of getting it/s and lass loss rate (have not tested loss rate yet)
ecker commented on issue ecker/ai-voice-cloning#15 2023-02-20 22:47:18 +00:00
Colab: ModuleNotFoundError: No module named 'tortoise'

Right, I forgot I ditched some additional install scripts to make things less complex yesterday. One moment while I update it.

ecker commented on issue ecker/ai-voice-cloning#14 2023-02-20 20:27:08 +00:00
Should the training data folder contain spaces, train.py will interpret as arguments

Remedied in bacac6daea94addb3807e3bde265685c2ce59766. God only knows why Python splits arguments by spaces, regardless of wrapping it in quotes. I guess it makes sense why the issue only pops up…

ecker closed issue ecker/ai-voice-cloning#14 2023-02-20 20:27:08 +00:00
Should the training data folder contain spaces, train.py will interpret as arguments
ecker pushed to master at ecker/ai-voice-cloning 2023-02-20 20:23:29 +00:00
bacac6daea handled paths that contain spaces because python for whatever god forsaken reason will always split on spaces even if wrapping an argument in quotes
ecker commented on issue ecker/ai-voice-cloning#14 2023-02-20 19:57:06 +00:00
Should the training data folder contain spaces, train.py will interpret as arguments

I had a gut feeling it was going to crop up. I'm not necessarily sure how, since I thought I took care in wrapping the proxied argument in "s, but I guess that's not good enough. I'll replicate…

ecker closed issue ecker/ai-voice-cloning#7 2023-02-20 15:49:05 +00:00
Something went wrong
ecker commented on issue ecker/ai-voice-cloning#6 2023-02-20 15:48:45 +00:00
Error when running start.bat

Given the revamped configuration generation process and (attempting to) unload all models on training, you might be able to get it working again if you lower the Mega Batch Factor or whatever…

ecker closed issue ecker/ai-voice-cloning#3 2023-02-20 15:46:52 +00:00
FileNotFoundError immediately after starting training
ecker commented on issue ecker/ai-voice-cloning#13 2023-02-20 15:46:30 +00:00
Difference

Since the project has grown in scope, mrq/tortoise-tts has been restructured to only provide my changes to tortoise-tts itself (only generating), while mrq/ai-voice-cloning is more about the one…

ecker closed issue ecker/ai-voice-cloning#12 2023-02-20 15:32:25 +00:00
Trying to train but I get "AttributeError: 'NoneType' object has no attribute 'transcribe'"
ecker commented on issue ecker/ai-voice-cloning#12 2023-02-20 15:32:19 +00:00
Trying to train but I get "AttributeError: 'NoneType' object has no attribute 'transcribe'"

Fixed in 37ffa60d1459fde4e1749d6e30980ab0178cc10d. Sorry for the inconvenience.

ecker pushed to master at ecker/ai-voice-cloning 2023-02-20 15:31:50 +00:00
37ffa60d14 brain worms forgot a global, hate global semantics
ecker commented on issue ecker/ai-voice-cloning#12 2023-02-20 15:28:42 +00:00
Trying to train but I get "AttributeError: 'NoneType' object has no attribute 'transcribe'"

Guess I didn't manage to get around testing transcribing if I broke it. One moment while I fix it.

ecker commented on issue ecker/ai-voice-cloning#11 2023-02-20 15:27:21 +00:00
Cuda out of memory

So the source files are 3 WAV files, that combined are less than 1 MB so very small. And all together it has about 5~6 sentences Trained on that for about 20 mins and generated a few…

ecker pushed to master at ecker/ai-voice-cloning 2023-02-20 00:21:29 +00:00
d17f6fafb0 clean up, reordered, added some rather liberal loading/unloading auxiliary models, can't really focus right now to keep testing it, report any issues and I'll get around to it
ecker pushed to master at ecker/ai-voice-cloning 2023-02-19 23:29:25 +00:00
c99cacec2e oops
ecker commented on issue ecker/ai-voice-cloning#11 2023-02-19 23:26:19 +00:00
Cuda out of memory

I checked through your issues page, but I'm not actually sure how to solve it even with what was written. I added the \n to "Line delimiter", the error suggests changing max_split_size_mb but I…