ecker
18 Followers · 0 Following

  • 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
Repositories
10
 Projects Packages Public Activity Starred Repositories
ecker pushed to master at ecker/ai-voice-cloning 2023-03-16 20:48:11 +00:00
2ca623046e setup bnb on windows as needed
ecker commented on issue ecker/ai-voice-cloning#133 2023-03-16 18:44:22 +00:00
Mispronouncing certain letters for slavic languages

That's with the ipa.json from 1a8c5de517 and the existing diffusion model?

Correct.

I tried again, this time with aggressive settings (LR: 0.0001, schedule: [2,4,9], and not…

ecker commented on issue ecker/ai-voice-cloning#133 2023-03-16 17:11:16 +00:00
Mispronouncing certain letters for slavic languages

Tried again, added some missing IPAs and merges; sounds like James Sunderland if he was Fr*nch: https://vocaroo.com/1c9ylHLoNrwU ("You're not Mary." => jʊɹ nɑːt mɛɹi.). Seeing the mapping…

ecker pushed to master at ecker/ai-voice-cloning 2023-03-16 15:33:30 +00:00
c4edfb7d5e unbump rocm5.4.2 because it does not work for me desu
ecker pushed to master at ecker/ai-voice-cloning 2023-03-16 15:30:27 +00:00
6ad02ccd0d unbump rocm5.4.2 because it does not work for me desu
ecker pushed to master at ecker/ai-voice-cloning 2023-03-16 15:09:18 +00:00
520fbcd163 bumped torch up (CUDA: 11.8, ROCm, 5.4.2)
ecker pushed to master at ecker/ai-voice-cloning 2023-03-16 14:59:17 +00:00
1a8c5de517 unk hunting
ecker pushed to master at ecker/ai-voice-cloning 2023-03-16 14:58:37 +00:00
887def32f1 unk hunted
ecker pushed to master at ecker/ai-voice-cloning 2023-03-16 14:41:50 +00:00
46ff3c476a fixes v2
ecker pushed to master at ecker/DL-Art-School 2023-03-16 14:41:25 +00:00
0db8ebc543 deduce if preprocessing text by checking the JSON itself instead
ecker pushed to main at ecker/tortoise-tts 2023-03-16 14:41:15 +00:00
af78e3978a deduce if preprocessing text by checking the JSON itself instead
ecker pushed to master at ecker/tortoise-tts 2023-03-16 14:41:11 +00:00
af78e3978a deduce if preprocessing text by checking the JSON itself instead
ecker pushed to master at ecker/ai-voice-cloning 2023-03-16 14:24:54 +00:00
0408d44602 fixed reload tts being broken due to being as untouched as I am
ecker pushed to master at ecker/ai-voice-cloning 2023-03-16 14:23:50 +00:00
aeb904a800 yammed
ecker commented on issue ecker/ai-voice-cloning#143 2023-03-16 14:21:27 +00:00
Error when preparing dataset

Should be fixed in commit f9154c4db1869c12904478d35bea2f449f9f8880.

ecker commented on issue ecker/ai-voice-cloning#144 2023-03-16 14:21:09 +00:00
ValueError when generating prompt

Should be fixed in mrq/tortoise-tts commit e201746eeb.

I'm still not sure how it wasn't erroring out for me.

ecker pushed to master at ecker/ai-voice-cloning 2023-03-16 14:20:13 +00:00
f9154c4db1 fixes
ecker pushed to main at ecker/tortoise-tts 2023-03-16 14:19:42 +00:00
e201746eeb added diffusion_model and tokenizer_json as arguments for settings editing
ecker pushed to master at ecker/tortoise-tts 2023-03-16 14:19:40 +00:00
e201746eeb added diffusion_model and tokenizer_json as arguments for settings editing
ecker commented on issue ecker/ai-voice-cloning#133 2023-03-16 14:10:38 +00:00
Mispronouncing certain letters for slavic languages

With the default tokenizer, yea. It just strips out all the accents.

That's because DLAS/tortoise's VoiceBpeTokenizer will preprocess all text:

  text = convert_to_ascii(text)
  text =…