Fun-ASR-Realtime released with one model supporting 30 languages and 16 dialects, streaming accuracy near offline levels
Original title:🔥Fun-ASR-Realtime发布:语音识别再升级!
AI summary
Tongyi Lab released Fun-ASR-Realtime, a single model supporting 30 languages and 16 dialects that achieves 87.8% semantic accuracy on the industrial-grade inhouse dialect benchmark. The model supports dynamic injection of hotwords and dialogue context for semantic disambiguation, keeps first-token latency at the hundred-millisecond level, delivers streaming recognition accuracy close to offline levels, and supports seamless multilingual switching. The API is now available on the Alibaba Cloud Bailian platform.
Key Takeaways
- Tongyi Lab released Fun-ASR-Realtime,a single model supporting 30 languages and 16 dialects that…
- 8% semantic accuracy on the industrial-grade inhouse dialect benchmark.
- The model supports dynamic injection of hotwords and dialogue context for semantic…
- The API is now available on the Alibaba Cloud Bailian platform.
Quick Facts
- Published
- Updated
- Source
- 公众号:通义实验室(千问)
- Category
- Model Releases
- AI score
- 65
多语言方言的流式识别将语音交互的可用边界从英语推向了东亚与方言市场,使原先因识别不准而难以落地的本地化客服、语音助手等应用获得了新的技术基础。
Related entities
Source & copyright
This article is based on curated titles and AI summaries from the AIHOT public API, with attribution and links back. Original copyright remains with the original publishers and authors; this site does not mirror full texts.
How to Cite
When citing this article, keep the title, date, and permalink, and credit Orange News as the source.
Orange News: Fun-ASR-Realtime released with one model supporting 30 languages and 16 dialects, streaming accuracy near offline levels. July 6, 2026. https://new.nat6.net/en/news/cmsdu3gxy014xrofzqejq2kuy/