mrq

18 Followers · 0 Following

• 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

Repositories

15

Projects Packages Public Activity Starred Repositories

mrq created branch master in mrq/tortoise-tts 2023-02-07 20:58:55 +00:00

mrq pushed to master at mrq/tortoise-tts 2023-02-07 20:58:55 +00:00

mrq pushed to main at mrq/tortoise-tts 2023-02-07 20:56:14 +00:00

6ebdde58f0 (finally) added the CVVP model weigh slider, latents export more data too for weighing against CVVP

mrq pushed to main at mrq/tortoise-tts 2023-02-07 18:35:46 +00:00

793515772a un-hardcoded input output sampling rates (changing them "works" but leads to wrong audio, naturally)

mrq pushed to main at mrq/tortoise-tts 2023-02-07 13:51:11 +00:00

5f934c5feb (maybe) fixed an issue with using prompt redactions (emotions) on CPU causing a crash, because for some reason the wav2vec_alignment assumed CUDA was always available

mrq commented on issue mrq/tortoise-tts#6 2023-02-07 13:50:10 +00:00

emotions not working with CPU

Should be fixed in commit ab3864563dd2a46a8fcbbd2bd39e4a523f587504. I say should, because, while I did force TorToiSe to use my CPU, I'm not sure how guaranteed it was, even with compute time…

mrq pushed to main at mrq/tortoise-tts 2023-02-07 13:46:43 +00:00

ab3864563d (maybe) fixed an issue with using prompt redactions (emotions) on CPU causing a crash, because for some reason the wav2vec_alignment assumed CUDA was always available

mrq closed issue mrq/tortoise-tts#5 2023-02-07 05:24:23 +00:00

In line 176 of app.py, I found it's very helpful to sort the voices in gradio as it can become quite difficult to find things otherwise

mrq pushed to main at mrq/tortoise-tts 2023-02-07 05:14:25 +00:00

d6b5d67f79 forgot to auto compute batch size again if set to 0

mrq pushed to main at mrq/tortoise-tts 2023-02-07 04:52:53 +00:00

66cc6e2791 changed ROCm pip index URL from 5.2 to 5.1.1, because it's what worked for me desu

mrq commented on issue mrq/tortoise-tts#5 2023-02-07 03:57:45 +00:00

In line 176 of app.py, I found it's very helpful to sort the voices in gradio as it can become quite difficult to find things otherwise

I feel like maybe I should have forked or made a pull request or something, but I've never used git, and this seems like quite a small brick to contribute to a pyramid.

No worries.

Added…

mrq pushed to main at mrq/tortoise-tts 2023-02-07 03:54:52 +00:00

6515d3b6de added shell scripts for linux, wrapped sorted() for voice list, I guess

mrq pushed to main at mrq/tortoise-tts 2023-02-07 00:26:43 +00:00

edd642c3d3 fixed combining audio, somehow this broke, oops

mrq pushed to main at mrq/tortoise-tts 2023-02-06 22:31:18 +00:00

be6fab9dcb added setting to adjust autoregressive sample batch size

mrq pushed to main at mrq/tortoise-tts 2023-02-06 21:43:44 +00:00

100b4d7e61 Added settings page, added checking for updates (disabled by default), some other things that I don't remember

mrq deleted branch dev from mrq/tortoise-tts 2023-02-06 18:46:01 +00:00

mrq pushed to main at mrq/tortoise-tts 2023-02-06 16:32:19 +00:00

240858487f Added encoding and ripping latents used to generate the voice

mrq pushed to main at mrq/tortoise-tts 2023-02-06 16:00:59 +00:00

92cf9e1efe Added tab to read and copy settings from a voice clip (in the future, I'll see about enmbedding the latent used to generate the voice)

mrq pushed to main at mrq/tortoise-tts 2023-02-06 14:18:05 +00:00

5affc777e0 added another (somewhat adequate) example, added metadata storage to generated files (need to add in a viewer later)

mrq pushed to main at mrq/tortoise-tts 2023-02-06 05:10:32 +00:00

b441a84615 added flag (--cond-latent-max-chunk-size) that should restrict the maximum chunk size when chunking for calculating conditional latents, to avoid OOMing on VRAM