 Mikei386
|
4cdd837482
|
Fix Seed-VC startup readiness and clarify Applio models
|
2026-09-09 21:29:06 +02:00 |
|
|
Mikei386
|
13a6714b2b
|
Add Seed-VC and Applio studio modes
|
2026-09-09 20:58:32 +02:00 |
|
|
Mikei386
|
51ed201f6c
|
Add optional 44.1 kHz X-VC enhancement
|
2026-09-09 17:21:20 +02:00 |
|
|
Mikei386
|
b4f4bf37fd
|
Document Athena containers and expand Hermes operator skill
|
2026-09-09 16:39:34 +02:00 |
|
|
Mikei386
|
87a2ae5704
|
Add X-VC voice conversion mode
|
2026-09-09 16:14:10 +02:00 |
|
|
Mikei386
|
17f1a08d7d
|
Document OmniVoice cloning gate
|
2026-09-09 15:15:10 +02:00 |
|
|
Mikei386
|
68d02f32bd
|
Add private Vevo2 voice studio mode
|
2026-09-09 13:38:00 +02:00 |
|
|
Mikei386
|
535bd751b5
|
Update llama.cpp runtime to build 10872
|
2026-09-09 12:27:36 +02:00 |
|
|
Mikei386
|
805228abfd
|
feat: add speech noise separation
|
2026-09-09 08:24:12 +02:00 |
|
|
Mikei386
|
0fd1966ae3
|
feat: add other stem extraction target
|
2026-09-09 07:48:26 +02:00 |
|
|
Mikei386
|
c5bbebecc6
|
fix: use ffmpeg export for all MP3 stem jobs
|
2026-09-09 01:03:39 +02:00 |
|
|
Mikei386
|
2342495d24
|
feat: add target-and-remainder stem extraction
|
2026-09-09 01:00:01 +02:00 |
|
|
Mikei386
|
0f0e77928a
|
fix: export Demucs stems through ffmpeg
|
2026-09-09 00:46:06 +02:00 |
|
|
Mikei386
|
e2f35517f8
|
feat: add multi-stem audio separation modes
|
2026-09-09 00:30:08 +02:00 |
|
|
Mikei386
|
30203bf13b
|
Fix dashboard separation mode switch
|
2026-09-08 23:08:58 +02:00 |
|
|
Mikei386
|
0069b61dbb
|
Add BS-RoFormer vocal separation mode
|
2026-09-08 19:23:29 +02:00 |
|
|
Mikei386
|
a4e894fe70
|
Add dual ACE-Step music studio interfaces
|
2026-09-08 18:46:54 +02:00 |
|
|
Mikei386
|
f58d61140e
|
Add persistent Athena music mode switching
|
2026-09-08 17:32:04 +02:00 |
|
|
Mikei386
|
56c382f71f
|
Persist ACE-Step XL-SFT quality defaults
|
2026-09-08 16:37:36 +02:00 |
|
|
Mikei386
|
eeebbd06eb
|
Fix UI networking across gateway recreation
|
2026-09-08 15:56:12 +02:00 |
|
|
Mikei386
|
5e18b7776b
|
Add ACE-Step XL SFT music experiment
|
2026-09-08 15:16:15 +02:00 |
|
|
Mikei386
|
2724861224
|
Speak aspect ratios correctly in TTS
|
2026-09-08 14:42:21 +02:00 |
|
|
Mikei386
|
f1ed51a302
|
Track tested models centrally
|
2026-09-08 14:26:46 +02:00 |
|
|
Mikei386
|
30fdbd4b7a
|
Document max-VRAM IQ3 speed tests
|
2026-09-08 14:14:46 +02:00 |
|
|
Mikei386
|
3e3fbbe9bd
|
Document profile-wide GSQ-RCO A-B test
|
2026-09-08 13:48:47 +02:00 |
|
|
Mikei386
|
edb845eb19
|
Document GSQ-RCO IQ3_S A-B results
|
2026-09-08 12:10:08 +02:00 |
|
|
Mikei386
|
f7ff14a1ce
|
Upgrade Beta 1 to GSQ-RCO IQ3_S
|
2026-09-08 11:46:52 +02:00 |
|
|
Mikei386
|
5746ac0e2c
|
Document removal of photo restoration experiment
|
2026-09-08 10:56:30 +02:00 |
|
|
Mikei386
|
e82e0340e4
|
Remove ineffective photo restoration pipeline
|
2026-09-08 09:40:43 +02:00 |
|
|
Mikei386
|
f82dc081c9
|
Disable text conditioning for faithful restoration
|
2026-09-08 08:52:49 +02:00 |
|
|
Mikei386
|
3220a67f1b
|
Deduplicate Hermes image references
|
2026-09-08 08:45:57 +02:00 |
|
|
Mikei386
|
636e48ce93
|
Merge restoration policy into system message
|
2026-09-08 08:37:53 +02:00 |
|
|
Mikei386
|
118e32005e
|
Add explicit HYPIR restoration profile
|
2026-09-07 22:52:59 +02:00 |
|
|
Mikei386
|
2ae61baec7
|
Prepare isolated Qwen3.8 IQ4_XS A/B tests
|
2026-09-07 16:22:51 +02:00 |
|
|
Mikei386
|
1844551534
|
Add dual-GPU FLUX 9B image pipeline
|
2026-09-07 15:42:45 +02:00 |
|
|
Mikei386
|
61aa20cb52
|
Add native Qwen3 TTS streaming for Hermes
|
2026-09-05 16:06:27 +02:00 |
|
|
Mikei386
|
5f3d064bb4
|
Smooth Hermes TTS sentence transitions
|
2026-09-05 15:28:11 +02:00 |
|
|
Mikei386
|
cbc312257c
|
Normalize German TTS abbreviations and units
|
2026-09-05 15:11:57 +02:00 |
|
|
Mikei386
|
63e7a93ee4
|
Improve German TTS range handling
|
2026-09-05 14:07:44 +02:00 |
|
|
Mikei386
|
cc416150a8
|
Replace XTTS with Qwen3-TTS
|
2026-09-05 13:30:14 +02:00 |
|
|
Mikei386
|
44e1c1c50e
|
Fix Hermes cron delivery to Docker profiles
|
2026-09-04 23:41:47 +02:00 |
|
|
Mikei386
|
f553108912
|
Add Qwen GSQ-RCO Beta 1 profile
|
2026-09-04 14:36:50 +02:00 |
|
|
Mikei386
|
744a207e5a
|
Stabilize XTTS sentence endings
|
2026-09-04 09:30:04 +02:00 |
|
|
Mikei386
|
8dac735680
|
Reduce XTTS tail hallucinations
|
2026-09-04 08:32:45 +02:00 |
|
|
Mikei386
|
be8a654f1e
|
Preserve sentence prosody in local TTS
|
2026-09-04 08:11:05 +02:00 |
|
|
Mikei386
|
5afdf46a7c
|
Compress hallucinated XTTS silence
|
2026-09-03 23:46:32 +02:00 |
|
|
Mikei386
|
435c59da41
|
Stream Athena Talk replies incrementally
|
2026-09-03 23:36:51 +02:00 |
|
|
Mikei386
|
384d81f6cd
|
Normalize question marks before local TTS
|
2026-09-03 23:24:15 +02:00 |
|
|
Mikei386
|
023c2ee40d
|
Smooth Talk playback and gate audio tail
|
2026-09-03 23:19:51 +02:00 |
|
|
Mikei386
|
65ce9642cc
|
Stabilize Athena Talk audio delivery
|
2026-09-03 23:11:44 +02:00 |
|