What is Seed Audio 1.0?

Seed Audio 1.0 is ByteDance Seed's all-in-one audio generation model for creating complete sound scenes. It uses text, image, or audio context to guide multi-speaker dialogue, emotional delivery, native accents, ambience, background music, and foley-style effects. It is positioned beyond plain text-to-speech, enabling layer-aware output with dialogue, BGM, ambience, and sound effects in one creative pass.

How to use Seed Audio 1.0?

To use Seed Audio 1.0 online, write a sound-scene prompt describing characters, language, emotion, location, dialogue, ambience, music direction, and sound events. Optionally add up to three audio references or one image reference. Choose output settings like voice behavior, format, sample rate, speed, volume, pitch, and credit cap. Then generate and review the audio in the SeedAudio.co workspace, starting with a short draft and refining as needed.

Core features of Seed Audio 1.0

All-in-one audio generation: dialogue, emotion, ambience, BGM, and foley in a single pass
Multimodal prompting: text, image, and audio context
Multi-speaker voice continuity for longer scenes
Emotion and accent control for natural delivery
Scene ambience and BGM generation
Longer audio scenes up to 2 minutes per generation
Layer-aware output with separate stems for dialogue, BGM, and ambience