
By Florian |
Mirelo Studio is built for creators who move at the speed of their ideas. Generate rich, synced sound in seconds and explore variants until it's just right — perfect for content creators, AI videos, and rapid prototyping!
Try it now -> mirelo.ai/studio
After months of beta testing (thank you to everyone who joined early!), Mirelo Studio is now live:
Under the hood, Mirelo Studio runs on Mirelo SFX 1.5: our most advanced video-to-sound effect model, engineered to make your visuals sound real.
This model update brings major improvements across every layer of performance:
In a blind pairwise comparison tests, independent raters* compared generations from Mirelo SFX 1.5 and the two most popular existing video-to-sfx alternatives (Kling Text-to-Audio and Tencent-Hunyuan VideoFoley). In average, Mirelo SFX achieved a total win rate of 68.3% (excluding ties) and 73.2% (including ties), clearly outperforming both alternatives.

Side-by-side examples
Competitors often struggle with unwanted sounds like music or speech (see, e.g., examples 1, 3 and 4 below). Mirelo SFX also avoids distortion and other audio artifacts introduced by competitors, creating realistic, high-quality effects across various scenes (see examples 2 through 4 below).
Try it now -> mirelo.ai/studio
Learn more about Mirelo SFX — the product for AI video-to-sound across Studio, plugins, and API.
* We recruited 15 external listeners from a crowd-sourcing platform via mabyduck. The raters were pre-screened by mabyduck to ensure they listen closely to the given audio. Raters performed pairwise comparisons between different audio generations for a given video, with the option to vote for a "tie". 555 pairwise comparisons were collected in total.