Unverified50% confidenceFactExact time
Live Transcribe最初于2019年发布，利用端侧+云端混合的语音识别技术将环境声音实时转为文字，支持超过80种语言
1
Sources
50%
Confidence
Long-term
Relevance
8/14/2026
First Seen
Sources
Related Claims
VerifiedWhisper is an automatic speech recognition model open-sourced by OpenAI in 2022, built on a Transformer architecture and trained on 680,000 hours of multilingual audio data.72% similarUnverifiedByteDance has released Seed Live Interpret 2.0, an end-to-end simultaneous interpretation model that approaches human-level latency and accuracy for Chinese-English simultaneous interpretation and supports zero-shot voice cloning.68% similarUnverifiedCodec-based approaches (such as VALL-E and SoundStorm) encode speech into discrete tokens and then generate output using a language model approach, achieving better zero-shot voice cloning results.66% similarUnverifiedAlibaba Qianwen's real-time speech recognition model FoneASR Realtime has been upgraded to support 16 dialects and 30 languages66% 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66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/746679API
curl https://kongchang.com/api/v1/knowledge/claims/746679MCP
get_claim(id=746679)