Record the current voice service readiness and remaining blocker

Constraint: the docs must reflect the real runtime state after installing CPU torch: health is up, but end-to-end voice recognition still times out
Rejected: declaring the voice API complete | the current smoke still does not return a final recognition payload within the timeout window
Confidence: medium
Scope-risk: narrow
Directive: keep status docs synchronized with actual smoke results, especially partial readiness states
Tested: /usr/local/miniconda3/bin/python -m unittest discover -s acr-engine/tests -v; /health endpoint returns ok under uvicorn; direct /recognize/voice smoke currently times out after CPU torch install
Not-tested: successful end-to-end /recognize/voice result payload within timeout

@@ -9,7 +9,8 @@
 
 Fresh evidence:
 - `/usr/local/miniconda3/bin/python -m unittest discover -s acr-engine/tests -v` => `Ran 7 tests, OK`
-- 当前环境缺 `uvicorn`，服务 smoke 尚不能直接启动，需要先补运行依赖。
+- `/usr/local/miniconda3/bin/python -m uvicorn src.service.app:app --host 127.0.0.1 --port 8000` 下 `/health` 已返回 `ok`
+- architect review approved the FAISS/local + voice-query direction, with the next unblocker remaining `torch` / full voice inference smoke.
 
 
 ## 2026-06-03 20-song local ACR workflow in acr-engine

@@ -24,9 +24,10 @@ flowchart TD
 | benchmark report 已生成 |  |
 | model card 已生成 |  |
 | license registry 已更新 |  |
-| service smoke test 通过 | partial: `/health` OK, `/recognize/voice` blocked by missing `torch` |
+| service smoke test 通过 | partial: `/health` OK, CPU torch installed, `/recognize/voice` end-to-end smoke still times out |
 | dataset whitelist 已确认 |  |
 | changelog 已更新 | yes |
+| architect review completed | yes (approved with watch) |
 
 ## 3. 文字说明
 - 任何缺失项都不能视作商用可发布

@@ -29,8 +29,8 @@
 - FastAPI 服务当前状态：
   - `acr-engine/src/service/app.py` 已新增 `POST /recognize/voice`
   - `/health` 可正常启动并返回 `ok`
-  - 当前 `POST /recognize/voice` 仍被 `torch` 缺失阻塞，返回清晰错误：
-    - `Engine dependencies unavailable: No module named 'torch'`
+  - architect review: approved with watch；当前 split（本地 FAISS / 可选 ChromaDB / 生产 pgvector）方向成立
+  - 当前 `POST /recognize/voice` 仍需完整 smoke 复核；此前明确阻塞点是 `torch` 缺失，现已补 CPU 版 `torch`，当前剩余现象是 voice 端到端 smoke 超时，需要继续收敛推理耗时或缓存策略
 - 当前 docs 已做第一轮简化：
   - `docs/README.md` 只保留最新架构与最短阅读顺序