88Z AI, 16 July — Alibaba Tongyi Lab releases Qwen-Audio-3.0-Realtime, supporting emotional tone adjustment, voice cloning, and background filtering, ranking first in comprehensive voice reasoning evaluations.
At a glance
- Supports contextual emotion adjustment, voice cloning, and voiceprint-level background filtering
- Ranks first overall in Artificial Analysis voice reasoning evaluation
The update
Alibaba Tongyi Lab has officially launched Qwen-Audio-3.0-Realtime, previously known as Fun-Realtime-AudioChat. This new model features strong context awareness, dynamically adjusting tone and emotion based on conversational context, while supporting high-precision voice cloning. It also introduces voiceprint-level background filtering to purify audio input, ensuring clear interaction even in noisy environments. Leveraging multimodal duplex dialogue technology, the model enables natural and fluid real-time communication. In the voice reasoning sub-category evaluation by Artificial Analysis, Qwen-Audio-3.0-Realtime achieved outstanding results, ranking first overall and surpassing GPT-Realtime-2, marking a significant breakthrough in real-time voice interaction technology.
Sources & editorial note
This article is an AI-assisted synthesis by 88Z AI based on the public sources below. Product claims and availability are attributed to their publishers. It does not represent independent testing or first-hand reporting.
- 通义实验室Source published
Source dates above refer to the original announcements; the date under the headline is this article’s publication on 88Z AI.