Model releaseModerate global significanceConfirmed confidence
ByteDance Seed releases Seeduplex
ByteDance Seed releases the native full-duplex voice model Seeduplex and rolls it out across the Doubao app.
What happened
Event details
Seeduplex uses a native end-to-end framework that processes listening and speaking in parallel. It continuously interprets the user's audio and decides whether to listen, wait or respond in situations involving noise, side conversations, pauses and interruptions.
Assessment
Why it matters
A published model record with a documented capability or availability change.
68/100Global significance score. Regional effects are recorded only when the evidence supports a meaningful difference.
Availability
Access notes
Available across the Doubao app at launch.