AI 导读
另一个我觉得在架构上很有意思的细节：它是一个单一的 Audio Encoder + LLM 基础架构，同时服务于离线和流式识别。 一个基座，可配置的 chunk 大小，离线侧没有精度损失。 实际上这意味着你不需要训练和维护两个会随时间漂移的模型，而是在一条延迟/质量轴上调优一个系统。
正文
Another detail that I find architecturally interesting: it's a single Audio Encoder + LLM foundation serving both offline and streaming recognition.
One base, configurable chunk sizes, and no accuracy tax on the offline side.
In practice that means you're not training and maintaining two models that drift apart over time, you're tuning one system along a latency/quality axis.
来源：@rohanpaul_ai · x.com