Marian Basti
Marian Basti
the result obtained by eval_model or synthesis is much worse than which is obtained by train process
What datasets and presets are you using?
the result obtained by eval_model or synthesis is much worse than which is obtained by train process
What frontend selected? I'm trying to train on spanish speakers and the results are a litte gibberish, but not noise.
the result obtained by eval_model or synthesis is much worse than which is obtained by train process
Shouldn't be so noisy. This is what i get with 40000 steps on 13 speaker dataset.  es frontend, so no phonetics dictionary
the result obtained by eval_model or synthesis is much worse than which is obtained by train process
Please let me know how well it goes with that batch size
I got it running on Windows 10 x64 and python 3.7
On further inspection, the module doesn't seem to have any of the functions or attributes it should have  I'm sure something went wrong on the build.
You could set it up as to when asked for a, say, 0.25 interpolation, first do a normal one and then interpolate the generated frame with the first one, and...
New commit! [https://github.com/AUTOMATIC1111/stable-diffusion-webui/commit/5a6387e189dc365c47a7979b9040d5b6fdd7ba43#diff-c6510edc055cb6d0f00bfe88ea1b559d06646e9600fed68c7433ea5efed91bc4](url)
Thank for the suggested repo! Its got plenty of nice features, but for the moment i was looking for OpenAI compatible one, and yours fits perfectly. I made the API_KEY...
https://github.com/NVlabs/Sana/pull/177 is working on this