Refresh the handoff docs for the voice-query ACR path

Constraint: the handoff should reflect the current FAISS-first local workflow and the partially wired voice service without claiming end-to-end inference is ready
Rejected: waiting for full torch-backed service completion before documenting progress | would hide the current repo state and block clean session handoff
Confidence: high
Scope-risk: narrow
Directive: keep future handoff updates focused on what is runnable now, what is blocked, and the next shortest path to unblock it
Tested: /usr/local/miniconda3/bin/python -m unittest discover -s acr-engine/tests -v; /usr/local/miniconda3/bin/python -m uvicorn src.service.app:app --host 127.0.0.1 --port 8000 with successful /health response
Not-tested: successful /recognize/voice inference until torch is installed

@@ -5,6 +5,47 @@
 
 ## 一页结论
 
+### 最新交付快照（2026-06-03 本地 20-song + voice pipeline）
+
+- 当前样本策略已明确分层：
+  - `/workspace` 仅作为样本/素材来源
+  - `acr-engine/` 才是训练、检索、评测、服务主工程
+- 当前本地 20-song 验证路线已固定：
+  - 本地小样本优先 **FAISS**
+  - `chromadb` 作为可选对照后端
+  - 生产环境继续保留 **pgvector**
+- 新增本地 20-song 入口：
+  - `acr-engine/scripts/local_music20_acr.py`
+  - 结果产物：`acr-engine/data/local_eval/music20_summary.json`
+- 当前本地 20-song 结果：
+  - `type_1 -> type_11`: `top1=1.0`, `top3=1.0`
+  - `type_7 -> type_11`: `top1=0.45`, `top3=0.65`
+  - `type_8 -> type_11`: `top1=0.4667`, `top3=0.7333`
+  - `type_16 -> type_11`: `top1=0.4167`, `top3=0.4167`
+- 新增哼唱/录音接入基础能力：
+  - `acr-engine/src/data/voice_chunker.py`：voice -> chunk
+  - `acr-engine/scripts/build_humming_eval_manifest.py`：chunk -> `humming_real` manifest
+  - `acr-engine/src/utils/context_exporter.py`：命中 reference window -> 上下文 clip（默认 10s）
+- 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'`
+- 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
+
+新 session 最短启动路径（当前推荐）：
+1. 读 `docs/session-handoff.md`
+2. 读 `docs/README.md`
+3. 读 `docs/CHANGELOG.md`
+4. 直接从 `acr-engine/README.md` 与 `acr-engine/scripts/local_music20_acr.py` 继续
+
+当前最重要的下一步：
+1. 给当前解释器安装/修复 `torch`
+2. 真正跑通 `POST /recognize/voice`
+3. 把哼唱评测集接入 `evaluate.py` 或独立评测脚本
+4. 继续做 docs 第二轮收敛，只保留当前有效主文档
+
 ### 最新交付快照（2026-06-02 16:12 UTC）
 
 - 当前状态：先交付，后续重启继续