Unverified50% confidenceFactExact time
XiaoWu采用本地LLM作为后处理层，对语音识别输出进行校对和润色
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026
Sources
XiaoWu Voice Input Method: Powered by Local LLMs for Accurate Offline Speech Recognition
bilibilii-xiaowu6/23/2026
Related Claims
UnverifiedXiaoWu是一款基于本地大语言模型的语音输入法，支持完全离线运行76% similarUnverified讯飞语音引擎在中文语音识别方面准确率高，中译外场景表现优于外译中的OCR扫描场景69% similarUnverified语音编码器 + LLM 解码器的组合是 2023 年以来语音大模型的主流范式，Qwen-Audio、Gemini Audio、Phi-4-audio 等均采用类似设计68% similarVerified语音识别技术经历了从隐马尔可夫模型（HMM）到循环神经网络（RNN）再到端到端模型的演进67% similarUnverifiedAI录音笔的转写加总结能力本质上是将语音识别ASR技术与大语言模型LLM串联部署的产物67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/41194API
curl https://kongchang.com/api/v1/knowledge/claims/41194MCP
get_claim(id=41194)