Minimax H3 Local Generation: Character Dialogue and Lip Sync Test
TigerClaw305 · reddit · 2026-08-20
User shared a video generated locally using Comfy UI and Minimax H3. The video features a dialogue scene between two anthropomorphic characters in a sci-fi lounge. The post includes the full prompt, character reference images, and audio references. The user achieved a match between character visuals and specific voice timbres through detailed instructions.
More from Multimodal
- Minimax H3 Workflow Fixes Distant Faces Without Latent Upscaling — Support_Marmoset · 2026-08-20
- Integrating Minimax H3 into YouTube Video Production Pipeline — scsonicshadow · 2026-08-20
- Hundreds of 2D photos turned into a flyable 3D spatial memory palace — bilawalsidhu · 2026-08-20
- YouArt introduces depth map feature with maximum precision and minimal cost — Scobleizer · 2026-08-20
- Local RTX 3090 Test: Head Swapping with Minimax Inpainting — Repulsive-Rush3505 · 2026-08-20
- NORD SOMMAR: A 4-min Swedish BL micro-drama created with Kling + Flux — ReallyNiceShipper · 2026-08-20