Volcano Engine launches Doubao audio model 1.0 for end-to-end audio creation

Volcano Engine just launched the Doubao audio generation model 1.0 (Doubao-Seed-Audio 1.0), which for the first time supports reference generation. You can input text or audio to create target audio end-to-end. It keeps multiple characters' voices consistent over longer audio, saving editing time. With a single prompt, you can set character dialogues, emotions, background music, and atmosphere to produce a complete audio piece, replacing the old workflow of mixing separate tracks. The API is now available for beta testing, and individual users get 30 minutes of free creation on the Volcano Engine experience center. It will also come to apps like CapCut, Jimeng, and Fanqie.