Runway adds native audio generation to all paid plans
By Steven Van ·
Seed Audio 1.0 generates speech, sound design and music from text prompts, up to two minutes per generation.
Runway, long a video-first platform, just added sound: Seed Audio 1.0 is now live for all paid plans, generating speech, sound design, and music from text.
Speech, SFX, and music from one prompt
Seed Audio 1.0 spans the full audio stack — dynamic speech, sound design, and music — all from simple text prompts. Instead of stitching together separate tools for voiceover, ambience, and score, you describe what you want and get audio back.
Up to two minutes per generation
It produces up to 120 seconds in a single generation — enough for a full scene's narration, a music bed, or a layered soundscape, rather than short clips you have to chain together.
Closing Runway's audio gap
Runway built its name on generative video; native audio lets a project move from generated visuals to a finished, sounded cut without leaving the tool. Rolling it out to every paid plan — not just a top tier — makes it part of the default workflow.
Sources: Runway.