• 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#144 2023-03-16 13:46:39 +00:00
ValueError when generating prompt

I'll push a fix to mrq/tortoise-tts when I get a chance. Odd it didn't turn up in my many generation tests over the past few days.

You might also just need to update tortoise itself with the…

ecker commented on issue ecker/ai-voice-cloning#143 2023-03-16 13:44:53 +00:00
Error when preparing dataset

Temporary fix is to go into Settings and click on the default Tokenizer (./modules/tortoise/tortoise-tts/data/tokenizer.json or something).

I'll push a fix when I get a chance.

ecker pushed to master at ecker/ai-voice-cloning 2023-03-16 05:14:20 +00:00
ecker pushed to master at ecker/ai-voice-cloning 2023-03-16 04:54:20 +00:00
ecker commented on issue ecker/ai-voice-cloning#141 2023-03-16 04:53:05 +00:00
Training ETA seems way off

Oops, I found the problem. DLAS originally had the iteration rate divided by the batch size, so the web UI compensated by multiplying the reported iteration back up.

I forgot I had DLAS report…

ecker closed issue ecker/ai-voice-cloning#141 2023-03-16 04:53:05 +00:00
Training ETA seems way off
ecker pushed to master at ecker/ai-voice-cloning 2023-03-16 04:51:40 +00:00
4ac43fa3a3 I forgot I undid the thing in DLAS
ecker pushed to master at ecker/DL-Art-School 2023-03-16 04:37:55 +00:00
730f56aa87 some day I'll get a commit right on the first try
ecker pushed to master at ecker/ai-voice-cloning 2023-03-16 04:35:23 +00:00
da4f92681e oops
ecker pushed to master at ecker/tortoise-tts 2023-03-16 04:33:12 +00:00
1f674a468f added flag to disable preprocessing (because some IPAs will turn into ASCII, implicitly enable for using the specific ipa.json tokenizer vocab)
ecker pushed to main at ecker/tortoise-tts 2023-03-16 04:33:11 +00:00
1f674a468f added flag to disable preprocessing (because some IPAs will turn into ASCII, implicitly enable for using the specific ipa.json tokenizer vocab)
ecker pushed to master at ecker/ai-voice-cloning 2023-03-16 04:25:37 +00:00
ee8270bdfb preparations for training an IPA-based finetune
ecker pushed to master at ecker/DL-Art-School 2023-03-16 04:24:36 +00:00
730a04708d added flag to disable preprocessing (because some IPAs will turn into ASCII, implicitly enable for using the specific ipa.json tokenizer vocab)
ecker pushed to main at ecker/tortoise-tts 2023-03-16 04:24:24 +00:00
99618694db added flag to disable preprocessing (because some IPAs will turn into ASCII, implicitly enable for using the specific ipa.json tokenizer vocab)
ecker pushed to master at ecker/tortoise-tts 2023-03-16 04:24:20 +00:00
99618694db added flag to disable preprocessing (because some IPAs will turn into ASCII, implicitly enable for using the specific ipa.json tokenizer vocab)
ecker pushed to master at ecker/ai-voice-cloning 2023-03-16 04:20:15 +00:00
ecker pushed to master at ecker/DL-Art-School 2023-03-16 04:14:00 +00:00
3e14bbb09a added flag to disable preprocessing (because some IPAs will turn into ASCII, implicitly enable for using the specific ipa.json tokenizer vocab)
ecker pushed to main at ecker/tortoise-tts 2023-03-16 04:13:47 +00:00
658d8dab8a added flag to disable preprocessing (because some IPAs will turn into ASCII, implicitly enable for using the specific ipa.json tokenizer vocab)
ecker pushed to master at ecker/tortoise-tts 2023-03-16 04:13:44 +00:00
658d8dab8a added flag to disable preprocessing (because some IPAs will turn into ASCII, implicitly enable for using the specific ipa.json tokenizer vocab)
ecker pushed to master at ecker/ai-voice-cloning 2023-03-16 04:08:01 +00:00
4763acf849 preparations for training an IPA-based finetune