• 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
mrq pushed to main at mrq/tortoise-tts 2023-02-05 05:51:25 +00:00
a19060f765 Fixed accidentally not passing user-provided samples/iteration values (oops), fixed error thrown when trying to write unicode because python sucks
mrq pushed to main at mrq/tortoise-tts 2023-02-05 05:47:59 +00:00
f04e303780 Fixed accidentally not passing user-provided samples/iteration values (oops)
mrq deleted tag v2.3 from mrq/tortoise-tts 2023-02-05 04:35:34 +00:00
mrq deleted tag v2.2 from mrq/tortoise-tts 2023-02-05 04:35:32 +00:00
mrq deleted tag v2.1 from mrq/tortoise-tts 2023-02-05 04:35:29 +00:00
mrq commented on issue mrq/tortoise-tts#3 2023-02-05 04:32:59 +00:00
Suggestion: Snap the "Candidates" slider to integer values

Added in commit cab32e1f45c1933eee3c888232c9aeff71022adb ; desu I don't think I ever actually used the Candidates slider to see it defaults as a float, my bad.

I'm also in the process of…

mrq pushed to main at mrq/tortoise-tts 2023-02-05 04:27:51 +00:00
cab32e1f45 Forgot to add steps=1 to Candidates slider
mrq pushed to main at mrq/tortoise-tts 2023-02-05 04:18:45 +00:00
84a9758ab9 Set transformer and model folder to local './models/' instead of for the user profile, because I'm sick of more bloat polluting my C:\
mrq pushed to main at mrq/tortoise-tts 2023-02-05 03:54:13 +00:00
d29ba75dd6 cleaned up element order with Blocks, also added preset updating the samples/iterations counts
mrq pushed to main at mrq/tortoise-tts 2023-02-05 01:28:59 +00:00
ed33e34fcc Added choices to choose between diffusion samplers (p, ddim)
mrq pushed to main at mrq/tortoise-tts 2023-02-04 01:51:34 +00:00
5c876b81f3 Added small optimization with caching latents, dropped Anaconda for just a py3.9 + pip + venv setup, added helper install scripts for such, cleaned up app.py, added flag '--low-vram' to disable minor optimizations
mrq commented on issue mrq/tortoise-tts#1 2023-02-04 01:19:49 +00:00
Solving environment: failed with initial frozen solve.

I got a working setup with just Python 3.9 and pip in commit 0d8211e450044aaf240a15a33b65c7b940ced7f5 and updated the setup.bat to reflect it. I suppose it's not a true fix to the issue, but it…

mrq pushed to main at mrq/tortoise-tts 2023-02-04 01:17:29 +00:00
0d8211e450 Added small optimization with caching latents, dropped Anaconda for just a py3.9 + pip + venv setup, added helper install scripts for such, cleaned up app.py
mrq pushed to main at mrq/tortoise-tts 2023-02-04 01:15:19 +00:00
e8087e1c6b Added small optimization with caching latents, dropped Anaconda for just a py3.9 + pip + venv setup, added helper install scripts for such, cleaned up app.py
mrq commented on issue mrq/tortoise-tts#2 2023-02-03 21:47:03 +00:00
Running out of memory despite being able to run the other gardio version?

Reverted the patch in commit 8f20afc18f8fe3be36daee99be591d302d54ac1f for now. I'll try and see about cramming in a way to toggle between using those optimizations or not. It should be easy.

mrq pushed to main at mrq/tortoise-tts 2023-02-03 21:45:23 +00:00
8f20afc18f Reverted slight improvement patch, as it's just enough to OOM on GPUs with low VRAM
mrq commented on issue mrq/tortoise-tts#2 2023-02-03 21:35:20 +00:00
Running out of memory despite being able to run the other gardio version?

Rats. I suppose that's why it wasn't merged in the original repo. Some anon(s) mentioned a small performance…

mrq commented on issue mrq/tortoise-tts#1 2023-02-03 21:26:37 +00:00
Solving environment: failed with initial frozen solve.

any ideas?

I have a few:

mrq pushed to main at mrq/tortoise-tts 2023-02-03 04:56:51 +00:00
e8d4a4f89c Added progress for transforming to audio, changed number inputs to sliders instead
mrq pushed to main at mrq/tortoise-tts 2023-02-03 04:20:09 +00:00
ea751d7b6c forgot to copy the alleged slight performance improvement patch, added detailed progress information with passing gr.Progress, save a little more info with output