Qwen Large Model has upgraded its real-time speech recognition model to Fun-ASR-Realtime. This new model boasts a latency of under 100 milliseconds for the first character and achieves accuracy close to offline models. It supports recognition in 16 dialects and 30 languages. AI
IMPACT This upgrade enhances real-time speech processing capabilities, potentially improving user interaction and accessibility for AI applications.
RANK_REASON This is an upgrade to a specific component of a large language model, not a new frontier model release.
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
