MiniMax H3 voices sound too close and loud — user asks if an audio LoRA can fix the proximity problem
Dogluvr2905 · reddit · 2026-09-13
The author praises MiniMax H3 but flags a shared weakness: its generated voices (whether from reference audio or fully synthetic) are too loud and sound 'pasted in', as if the character is talking into a microphone a foot away, failing to sit acoustically in the scene. They've tried countless prompt combinations and feeding in low-volume reference audio — nothing works. The question: can a LoRA be trained to make H3's voices quieter and more 'distant' so they blend into the mix? A useful window into a real generation quirk: close-mic sound with no sense of distance or room ambience.
More from Multimodal
- Gemini 4 Carbon vs Opus 5.5 hands-on: 3x cheaper at $4.10 and twice as fast — vitaliychiley · 2026-10-31
- Surreal AI fairytale: a town obsessed with turning tears into yarn — gizakdag · 2026-10-31
- ComfyUI extension adds one-click saving of preview images with metadata — sitefall · 2026-10-31
- DeepMind scientist directs MKUltra microdrama with Gemini agents for ~$35 total — korymath · 2026-10-31
- Using a gamepad-controlled Godot camera path to drive Wan video generation — droidsculptor · 2026-10-31
- Customuse launches Decals to fix 3D AI models' blurry logos and text — SimplyAnnisa · 2026-10-31