Generate dialogue, music, ambience, and effects as one audio scene
SeedAudio 1.5 is an upcoming audio-generation model that creates dialogue, music, ambience, and sound effects together from a scene brief, addressing the work of assembling layered audio separately. It is aimed at creators producing video dubbing, podcasts, audiobooks, game scenes, ads, and other narrated or sound-designed content.
Published specifications describe 30 languages with regional variants, timestamp control, and video-aware dubbing. SeedAudio 1.5 is not live yet; the page’s current online generator runs Seed Audio 1.0.
Also launched this week
Comments
Sign in to leave a comment.