-
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
Oh yeah, I'll see what I can grab. For sure I'll grab the latest examples to test, and I'll load back the old models and do an eval to spit out audio with Vocos.
I've had the quarter-sized…
I'm really not too sure what to make of these results. There's some things that make sense in hindsight, but there's a lot of weird shit with this new run for the quarter-sized model at 1.5kbps /…
The first epoch has concluded for the fullsize model, I'm not too pleased since training definitely petered off, but at least it's still going down.  optimization to yaml bool, added support for converting/loading from an hdf5 dataset
mmmmmmm
For Russian, I can't imagine you needing to use a new tokenizer definition for Russian. If you want to be extra sure, I believe it's Utilities > Tokenizer, and you can type in…
If I remember right, the model merging code is shamelessly (but credited) lifted from voldy Automatic1111's Stable…
It being very, very general purpose for being able to do a variety of inputs + infilling seems very, very promising solely in the new model architecture (getting a bit tired of the typical…
Training the quarter sized model with the new dataset has stabilized. I don't know whether it's:
- skipping "validating" the phonemes, which requires loading every file several times and some…
can do 44.1KHz at 8kbps of bandwidth
Sob. I JUST quantized everything.
Training has resumed after a few days to transcribe the audiobooks + re-quantize the LibriTTS_R dataset. I think unironically, quantizing the audio through Encodec on my 4070Ti is much slower than…
Couldn't find any mention of them in the wiki or the issues
Whoops, I forgot to add it to the documentation, but a loose idea of it is in mrq/ai-voice-cloning#243…
If you have Embed Output Metadata enabled in settings, the latents used for that generation are "embedded" into the result sound file. You can take that into Utilities > Import / Analyze and…
I guess this is intented behaviour
"Intentional" in the sense there's no way to keep it loaded without it being a server such as with the main script, yes, it's intentional. There's no way to…
Sorry it took so long, especially for how simple the cli.py script ended up being. My work ethic has taken quite the nosedive.
Should be added in commit 76ed34ddd2e610d466cf2c6fb76e13eea6df27b…