Skip to content
AI News HubLIVE
Summary only

Alibaba Qwen Team Introduces Qwen3.5-LiveTranslate-Flash: Real-Time Multimodal Interpretation Across 60 Languages at 2.8-Second Latency

Alibaba's Qwen team has released Qwen3.5-LiveTranslate-Flash, a real-time multimodal translation model that processes audio and video simultaneously. The model covers 60 input languages and produces speech output in 29 languages at 2.8 seconds of latency. Key additions over the previous Qwen3 version include real-time speaker voice cloning, vision-enhanced comprehension via lip movements and on-screen text, and dynamic keyword configuration for domain-specific terminology. On FLEURS and CoVoST2 benchmarks, the model outperforms major commercial alternatives. It is available as an API-only model through Alibaba Cloud Model Studio using a WebSocket-based protocol.

SourceMarkTechPostAuthor: Asif Razzaq
Alibaba Qwen Team Introduces Qwen3.5-LiveTranslate-Flash: Real-Time Multimodal Interpretation Across 60 Languages at 2.8-Second Latency
Report an error

The correction channel is not available yet. You can copy the article reference below for later.

Correction instructions

Summary only

Under the source content permission, this page provides an editorial summary, key points, and a link to the original article.

Key points and analysis

Article intelligence

EngineersAdvanced

Key points

  • Supports 60 input languages and 29 speech output languages with a latency of just 2.8 seconds.
  • Introduces vision-enhanced comprehension using lip movements, gestures, and on-screen text to improve accuracy in noisy environments.
  • Real-time voice cloning preserves the speaker's original voice characteristics in the translated output.
  • Dynamic keyword configuration allows injection of domain-specific glossaries for reliable translation of technical terms.

Highlights and analysis are generated automatically and may contain errors. Check the original source.

More recent stories on this topic · Models