-
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
I whipped up a small script to play around with it and, while I had zero hitches actually getting it to run (which in post I guess I got lucky, apparently people have had it not work given the…
Oh, actually, there is a repo for it: https://github.com/serp-ai/bark-with-voice-clone
I'll play around with it. If it proves favorable then I guess I won't need VALL-E.
I guess I never sent my reply here, oops.
Would the trained models be useful for you to help test inference?
mmm, it shouldn't be necessary. I just need to spend half an hour at most to…
Have you seen https://github.com/Fictiverse/bark?
Seen it. I mentioned some thoughts here, but I'll mention my current thoughts: *…
Bit the bullet yesterday; transcribed the train-360 LibriTTS (sub)dataset, putting me at a total of 116 hours, 167520 lines (total, actual dataset could be more, but I dropped Westwood's Blade…
hope you've had the chance to get some rest
Hah.
I trained a model using enhuiz's version
Yeesh. I probably should have made training with the repo more palatable to save your time.…
Since I don't really got anywhere else to mention it, I think I squashed the error 500 bugs. I'm not sure why it happened recently, but fuck SystemD. I had to use coreadm in my global zone to…
The demos are pretty nice:
- it does that thing I remember TorToiSe doing where it'll have non-verbal utterances like slight lip smacks and breaths.
- it boasts having a model already pretrained,…
4f8dd0bb7a
I really need to spend a day to update the readme and make sure it actually reflects the repo
25c05677bc
pushing uncommited changes, something something modifying validation/evaluation pipeline, updated example yaml, some other things i can't remember
1de785950a
added note in README if your CUDA version is higher than your PyTorch CUDA version because updating is a detriment (I just want my Xorg working again)
mmm, yeah I definitely won't try and tackle that. I'll let the real experts deal with maturing it, and hopefully someone with the actual compute to play around with it and homebrewing a model.
F…
If I remember right, you can achieve voice "mixing" on an arbitrary number of voices by having all your source samples in the same folder. When the latents get computed, it'll all "average" out,…
Twice I had what I was going to say eaten away. I already spent maybe 45 minutes to an hour, so I'm keeping it as brief as I can. Apologies if it comes off rather curt.
Is the size the…
I'm biting the bullet and dumping in LibriTTS clean-100 (247 speakers, 30k unsegmented lines, don't have an idea about duration yet or final line count).
I'm getting really worried that I'm…
Are you doing a reset?
An LR/optimizer reset just discards the metadata used for training while retaining the actual model. This way, the iteration count is reset to zero, and the LR schedule…
09f47524c9
hotfix for deepspeed 0.9 (words cannot describe my disgust and irritation for it), also some uncommitted cleanups
I'll probe through them whenever I get the chance next. I incidentally worked on the dataset preparation process to be more cleaner and not a pain (stuff like fixing the phonemizer memory leak,…
faa8da12d7
modified logic to determine valid voice folders, also allows subdirs within the folder (for example: ./voices/SH/james/ will be named SH/james)