Doubao Real-time Voice Model 3.0 API Service Officially Launched
Published · Jun 18 · Thu Source · 火山引擎 (CN)

Doubao Real-time Voice Model 3.0 API Service Officially Launched

Volcano Engine has officially launched the Doubao Real-time Voice Model 3.0 (Seeduplex) API service and opened invitation-based testing. As a native full-duplex end-to-end voice large model, the model features three major advantages: precise adherence, anti-interference, and dynamic turn-taking detection. It supports calling tools to complete tasks during real-time conversations, achieving simultaneous listening, speaking, and task execution. Turn-taking detection latency has been reduced by approximately 250ms, and the interruption rate in complex scenarios has decreased by 40%.

KeywordsAPIDoubaoReal-timeVoiceModelServiceOfficiallyLaunched

Volcano Engine has officially launched the Doubao Real-time Voice Model 3.0 (Seeduplex) API service and opened invitation-based testing. As a native full-duplex end-to-end voice large model, the model features three major advantages: precise adherence, anti-interference, and dynamic turn-taking detection. It supports calling tools to complete tasks during real-time conversations, achieving simultaneous listening, speaking, and task execution. Turn-taking detection latency has been reduced by approximately 250ms, and the interruption rate in complex scenarios has decreased by 40%.

This page provides an editorial summary based on publicly available information. It is not a republished article. Use the source link below for the original report.