Live drop
ByteDanceDropped Jun 30, 20262 builds2 makers
One-prompt generation of mixed dialogue, music and SFX scenes with zero-shot voice cloning.
About this release
Seed Audio 1.0 is ByteDance’s universal audio generation model developed by the Seed research team. It produces natural-sounding speech, instrumental music, sound effects, and ambient audio together as finished tracks, supporting zero-shot voice cloning, multi-speaker dialogue, and editing operations such as extending clips or swapping lines. The model accepts text prompts along with optional reference audio or image guidance for consistent character voices and scene coherence. It was released on June 30, 2026, building on audio capabilities previously integrated in Seedance video models.
Read the official ByteDance launch pageWorth opening first
Paste the URL and Dropday.ai will credit you, track the build as it evolves, and put it on this drop.
Submit a build