blumia
21d68e2294
Add native streaming ASR: 8s-aligned audio slices are cached in the KV cache, only new audio is encoded, and the previous text is fed back as an assistant prefix with token rollback. Text matches offline results. Streaming keeps each step within the 2s budget for segments up to 20s. 新增原生流式解码：8s 对齐的音频切片存入 KV 复用，只编码新音频，并把 上一轮文本作为 assistant 前缀回喂（带 token 回滚）。转写结果与离线 一致；20s 以内的段落每步计算都在 2s 预算内。