-
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 incidentally had a similar issue with something unrelated in Python even with an en-US system. The "fix" was to set PYTHONUTF8=1.
I'll throw it into the start.bat and hope it doesn't have…
65a43deb9e
why didn't I also have it use chunks for computing the AR conditional latents (instead of just the diffusion aspect)
65a43deb9e
why didn't I also have it use chunks for computing the AR conditional latents (instead of just the diffusion aspect)
Running
./scripts/tortoise_tts.py
Beyond my scope, as:
- I haven't touched that file at all.
- I'm not supporting manually invoking tortoise-tts.
- my priority lies with [mrq/ai-voice-clo…
Noted, I'll make sure to add that as a note in the wiki.
dadb1fca6b
multichannel audio now report correct duration (surprised it took this long for me to source multichannel audio)
The recommendation now is to not use all-in-one files
Correct.
It's not the end of the world if you can't, but Whisper's timestamps aren't to be trusted as much as I had hoped.
make…
32d968a8cd
(disabled by default until i validate it working) added additional transcription text normalization (something else I'm experimenting with requires it)
66ac8ba766
added mel LR weight (as I finally understand when to adjust the text), added text validation on dataset creation
Anyways, the last thing I need to mend is how I'm "leveraging" the dataset by making it actually leverage the dataset, rather than something like "I see you segmented your audio, so I'll make some…
I forgot I was too fried over the weekend to respond.
What I'm wondering about is how to get segmentation right. It's surprising to hear that sentence by sentence is okay, since whisper has…
Allegedly WSL2 does support nccl, per NVIDIA's doc/blog/guide. I'm not too well-versed in how robust WSL2 is, but I imagine just using the…