37ae1be6d5Fixed spelling error in api.format_conditioning
lightmare
2023-02-07 19:46:02 +0000
f7274112c3un-hardcoded input output sampling rates (changing them "works" but leads to wrong audio, naturally)mrq2023-02-07 18:34:29 +0000
793515772aun-hardcoded input output sampling rates (changing them "works" but leads to wrong audio, naturally)mrq2023-02-07 18:34:29 +0000
55058675d2(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 availablemrq2023-02-07 07:51:05 -0600
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 availablemrq2023-02-07 07:51:05 -0600
328deeddaeforgot to auto compute batch size again if set to 0mrq2023-02-06 23:14:17 -0600
d6b5d67f79forgot to auto compute batch size again if set to 0mrq2023-02-06 23:14:17 -0600
6475045f87changed ROCm pip index URL from 5.2 to 5.1.1, because it's what worked for me desumrq2023-02-06 22:52:40 -0600
66cc6e2791changed ROCm pip index URL from 5.2 to 5.1.1, because it's what worked for me desumrq2023-02-06 22:52:40 -0600
5d76d47a49added shell scripts for linux, wrapped sorted() for voice list, I guessmrq2023-02-06 21:54:31 -0600
6515d3b6deadded shell scripts for linux, wrapped sorted() for voice list, I guessmrq2023-02-06 21:54:31 -0600
3b56c437aafixed combining audio, somehow this broke, oopsmrq2023-02-07 00:26:22 +0000
edd642c3d3fixed combining audio, somehow this broke, oopsmrq2023-02-07 00:26:22 +0000
a3c077ba13added setting to adjust autoregressive sample batch sizemrq2023-02-06 22:31:06 +0000
be6fab9dcbadded setting to adjust autoregressive sample batch sizemrq2023-02-06 22:31:06 +0000
d8c88078f3Added settings page, added checking for updates (disabled by default), some other things that I don't remembermrq2023-02-06 21:43:01 +0000
100b4d7e61Added settings page, added checking for updates (disabled by default), some other things that I don't remembermrq2023-02-06 21:43:01 +0000
d1172ead36Added encoding and ripping latents used to generate the voicemrq2023-02-06 16:32:09 +0000
240858487fAdded encoding and ripping latents used to generate the voicemrq2023-02-06 16:32:09 +0000
e25ec325feAdded 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)mrq2023-02-06 16:00:44 +0000
92cf9e1efeAdded 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)mrq2023-02-06 16:00:44 +0000
edb6a173d3added another (somewhat adequate) example, added metadata storage to generated files (need to add in a viewer later)mrq2023-02-06 14:17:41 +0000
5affc777e0added another (somewhat adequate) example, added metadata storage to generated files (need to add in a viewer later)mrq2023-02-06 14:17:41 +0000
b8b15d827dadded flag (--cond-latent-max-chunk-size) that should restrict the maximum chunk size when chunking for calculating conditional latents, to avoid OOMing on VRAMmrq2023-02-06 05:10:07 +0000
b441a84615added flag (--cond-latent-max-chunk-size) that should restrict the maximum chunk size when chunking for calculating conditional latents, to avoid OOMing on VRAMmrq2023-02-06 05:10:07 +0000
319e7ec0a6fixed up the computing conditional latentsmrq2023-02-06 03:44:34 +0000
a1f3b6a4dafixed up the computing conditional latentsmrq2023-02-06 03:44:34 +0000
3c0648beafupdated README (before I go mad trying to nitpick and edit it while getting distracted from an iToddler sperging)mrq2023-02-06 00:56:17 +0000
2cfd3bc213updated README (before I go mad trying to nitpick and edit it while getting distracted from an iToddler sperging)mrq2023-02-06 00:56:17 +0000
b23f583c4eForgot to rename the cached latents to the new filenamemrq2023-02-05 23:51:52 +0000
945136330cForgot to rename the cached latents to the new filenamemrq2023-02-05 23:51:52 +0000
c2c9b1b683modified how conditional latents are computed (before, it just happened to only bother reading the first 102400/24000=4.26 seconds per audio input, now it will chunk it all to compute latents)mrq2023-02-05 23:25:41 +0000
5bf21fdbe1modified how conditional latents are computed (before, it just happened to only bother reading the first 102400/24000=4.26 seconds per audio input, now it will chunk it all to compute latents)mrq2023-02-05 23:25:41 +0000
7b767e1442New tunable: pause size/breathing room (governs pause at the end of clips)mrq2023-02-05 14:45:51 +0000
8831522de9New tunable: pause size/breathing room (governs pause at the end of clips)mrq2023-02-05 14:45:51 +0000
ec31d1763aFix to keep prompted emotion for every split linemrq2023-02-05 06:55:09 +0000
c7f85dbba2Fix to keep prompted emotion for every split linemrq2023-02-05 06:55:09 +0000
9ef7f11c0aUpdated .gitignore (that does not apply to me because I have a bad habit of having a repo copy separate from a working copy)mrq2023-02-05 06:40:50 +0000
79e0b85602Updated .gitignore (that does not apply to me because I have a bad habit of having a repo copy separate from a working copy)mrq2023-02-05 06:40:50 +0000
98dbf56d44Skip combining if not splitting, also avoids reading back the audio files to combine them by keeping them in memorymrq2023-02-05 06:35:32 +0000
bc567d7263Skip combining if not splitting, also avoids reading back the audio files to combine them by keeping them in memorymrq2023-02-05 06:35:32 +0000
3e3634f36aFixed accidentally not passing user-provided samples/iteration values (oops), fixed error thrown when trying to write unicode because python sucksmrq2023-02-05 05:51:57 +0000
cd94cc8459Fixed accidentally not passing user-provided samples/iteration values (oops), fixed error thrown when trying to write unicode because python sucksmrq2023-02-05 05:51:57 +0000
26daca3dc6Forgot to add steps=1 to Candidates slidermrq2023-02-05 04:27:20 +0000
cab32e1f45Forgot to add steps=1 to Candidates slidermrq2023-02-05 04:27:20 +0000
111c45b181Set transformer and model folder to local './models/' instead of for the user profile, because I'm sick of more bloat polluting my C:\mrq2023-02-05 04:18:35 +0000
84a9758ab9Set transformer and model folder to local './models/' instead of for the user profile, because I'm sick of more bloat polluting my C:\mrq2023-02-05 04:18:35 +0000
d2aeadd754cleaned up element order with Blocks, also added preset updating the samples/iterations countsmrq2023-02-05 03:53:46 +0000
d29ba75dd6cleaned up element order with Blocks, also added preset updating the samples/iterations countsmrq2023-02-05 03:53:46 +0000
078dc0c6e2Added choices to choose between diffusion samplers (p, ddim)mrq2023-02-05 01:28:31 +0000
ed33e34fccAdded choices to choose between diffusion samplers (p, ddim)mrq2023-02-05 01:28:31 +0000
4274cce218Added small optimization with caching latents, dropped Anaconda for just a py3.9 + pip + venv setup, added helper install scripts for such, cleaned up app.py, added flag '--low-vram' to disable minor optimizationsmrq2023-02-04 01:50:57 +0000
5c876b81f3Added small optimization with caching latents, dropped Anaconda for just a py3.9 + pip + venv setup, added helper install scripts for such, cleaned up app.py, added flag '--low-vram' to disable minor optimizationsmrq2023-02-04 01:50:57 +0000
061aa65ac4Reverted slight improvement patch, as it's just enough to OOM on GPUs with low VRAMmrq2023-02-03 21:45:06 +0000
8f20afc18fReverted slight improvement patch, as it's just enough to OOM on GPUs with low VRAMmrq2023-02-03 21:45:06 +0000
4f359bffa4Added progress for transforming to audio, changed number inputs to sliders insteadmrq2023-02-03 04:56:30 +0000
e8d4a4f89cAdded progress for transforming to audio, changed number inputs to sliders insteadmrq2023-02-03 04:56:30 +0000
ef237c70d0forgot to copy the alleged slight performance improvement patch, added detailed progress information with passing gr.Progress, save a little more info with outputmrq2023-02-03 04:20:01 +0000
ea751d7b6cforgot to copy the alleged slight performance improvement patch, added detailed progress information with passing gr.Progress, save a little more info with outputmrq2023-02-03 04:20:01 +0000
aafef3a140Cleaned up the good-morning-sirs-dialect labels, fixed seed=0 not being a random seed, show seed on outputmrq2023-02-03 01:25:03 +0000
43f45274ddCleaned up the good-morning-sirs-dialect labels, fixed seed=0 not being a random seed, show seed on outputmrq2023-02-03 01:25:03 +0000
lightmare
mrq
Aymeric Weinbach
Aymeric Weinbach
Pooya Moradi
Dawid Laszuk
James Betker
chris
Bewatermy
Dawid
Remy
원빈 정