FM PRO TECH NEWS
Sonilo and generative media platform fal have launched Sound Effects 1.0, a new AI model designed to generate realistic sound effects directly from video or text.
Using video input, the model analyses on-screen movement, environments, scene context and timing before producing a finished audio track synchronised with the footage.
Creators can also generate standalone effects from written prompts, while optional instructions can guide the sound, emphasis or creative direction of a video-based result.
The system is designed to reduce the manual work involved in traditional sound design, where editors typically search libraries, place individual effects and align each sound with the correct frame.
Sound effects only work when they feel like they belong in the scene. Sound Effects 1.0 was built around that complete problem: understanding the footage, generating realistic audio, and synchronizing it automatically. – Trista Hong, Co-Founder of Sonilo
Sound Effects 1.0 supports videos of up to three minutes and is aimed at short-form content, advertising, gaming footage, product videos and narrative production.
fal will serve as the model’s exclusive API launch partner during its initial release, allowing developers to integrate video-conditioned sound generation into editing platforms, creator tools, games and wider generative media products.
The launch expands an existing partnership between the two companies. Sonilo Music v1.1 is already available through fal, offering video-to-music and text-to-music generation.
Together, the tools allow the same footage to guide both sound effects and music, creating a broader video-first audio workflow built around the timing, movement and emotional structure of an edit.
For FM PRO TECH, the bigger shift is from AI-generated assets to AI-assisted post-production. As visual generation improves, platforms are increasingly targeting the missing audio layer—reducing manual sound placement while moving closer to complete, production-ready media workflows.