HaiAI123

Curated Global AI Tools Directory

Industry update
Flat

Volcano Engine launches Doubao SeedRealtime real-time audio-video API

11/19/2026 — 11/19, 15:20·2 sources (all)·2 reports (all)

Key takeaways

  • Volcano Engine launched the API for Doubao SeedRealtime 1.0, a real-time audio-video call model. It unifies voice, visuals, timing, and expr…

Background & analysis

On November 19, 2026, Volcano Engine officially launched the API service for Doubao's real-time audio and video conversation model, SeedRealtime 1.0. According to a 36Kr report published at 14:50 that day, the model unifies sound, visuals, timing, and expression into a single end-to-end model, enabling real-time interaction on continuous multimodal streams so that it can watch, listen, and speak at the same time. Developers can now access the model through the API.

At 14:55 the same day, IT Home reported the launch as well and added details drawn from the company's own description. According to IT Home, compared with voice cascade setups that chain separate ASR, VLM, and TTS modules, SeedRealtime 1.0 reduces cumulative latency and information loss. Instead of waiting passively for user questions, the model keeps tracking changes in the scene and can speak up at the right moment, folding seeing, hearing, and speaking into one real-time decision system. The company also says the model natively fuses audio, visual, and temporal information, which helps resolve homophone ambiguity in context and understand temporal references in the visual stream. In crowded, noisy, open environments it can tell who is speaking and follow the surrounding context, making it suited to multi-person collaboration, cross-language guidance, and real-time question answering with visual context. The public record currently stops at the API launch, with no later version or feature update disclosed.

AI-generated from 2 reports · updated 3 days ago

Latest turnVolcano Engine has launched the API for Doubao SeedRealtime 1.0, an end-to-end model that unifies audio, video and timing so it can watch, listen and speak at the same time. The company says the single-model design avoids the latency stacking and information loss of chaining ASR, VLM and TTS modules, and lets the model tell speakers apart in noisy multi-person settings and speak up on its own.

Related AI tools

updream
24-hour heatHeat index 50 · peak 56 · 23h ago
24 hours agonow

Reports on this story headlines open the original

Nov 19
  1. Volcano Engine has launched the API for Doubao SeedRealtime 1.0, an end-to-end model that unifies audio, video and timing so it can watch, listen and speak at the same time. The company says the single-model design avoids the latency stacking and information loss of chaining ASR, VLM and TTS modules, and lets the model tell speakers apart in noisy multi-person settings and speak up on its own.

    IT之家AI score 65
  2. Volcano Engine has launched API access to SeedRealtime 1.0, Doubao's real-time audio-video calling model, 36Kr reported. The end-to-end model unifies voice, visuals, timing and expression in a single system, enabling real-time interaction over continuous multimodal streams so it can watch, listen and speak at once.

    36氪AI score 55

Other stories people are talking about

How is heat calculated?About the method

Heat counts how many independent sources covered a story in the last 48 hours: one source counts once no matter how many posts it published, decaying with a 24-hour half-life. What ranks first is what many people are talking about.

This page aggregates public feeds. Headlines and summaries are machine-organized and remain the property of the original authors; verify important facts at the source. If you believe a headline or summary infringes your rights, email the contact address in the footer with the page URL and basis of your claim, and we will remove or replace it after review.

Surge
Discussion rising fast
New
First report within 6 hours
Rising
Still gathering discussion

Back to AI industry updates →