Mirelo SFX 2.0 auto-designs full soundtracks for AI videos, no full regeneration needed
iamfakhrealam · x · 2026-10-21
Mirelo launched SFX 2.0, targeting a real pain point: since AI video models generate sound and video together, bad audio used to mean regenerating the whole clip.
The model watches your video, cuts music you never asked for, keeps dialogue, and builds a full sound design synced to on-screen action, consistent across clips of any length. Live today via API and on Runway and fal, with more platforms coming.
More from Multimodal
- NVIDIA demos interactive AI surgical simulator running on a single RTX PRO 6000 GPU — Scobleizer · 2026-10-21
- Topaz joins Adobe; Starlight Fast upscales 720p AI video to 4K/60fps — Mr_AllenT · 2026-10-21
- YuE2 Studio runs M-A-P's music model locally on 6GB VRAM with editable scores and LoRA training — wzwowzw0002 · 2026-10-21
- FLUX 3 Klein on RTX 3090: int8 Doubles Speed, but FLUX.2 klein 4B Wins on Quality — strata2signal · 2026-10-21
- EvaOmni: an evidence-grounded benchmark for audio-visual reasoning in omnimodal LLMs — VisionXLab · 2026-10-21
- WinARDet: compressed window autoregression stops box repetition loops in dense grounding — NankaiUniversity · 2026-10-21