添加双耳渲染功能

2026-09-06 19:14:24 +08:00
parent 329445ed25
commit 2a296099fa
38 changed files with 9660 additions and 214 deletions
@@ -0,0 +1,277 @@
+# Binaural rendering
+
+[中文](binaural.md) · [Back to README](../README.en.md)
+
+JustOneCacophony's binaural backend supports three HRTF sources:
+`SimpleFreeFieldHRIR` SOFA, the Rosella `.personalized_headphone` model exported
+by Dolby's official personalization scan (its JSON parsing is implemented by
+this project and invokes no Dolby software), and the `.jochrtf` cache compiled
+from SOFA. SOFA is compiled into an in-memory directional field when the model
+is loaded. A `.jochrtf` file is only a disposable, reproducible JOC compiled
+HRTF cache; it is neither an interchange format nor a prerequisite for using
+SOFA.
+
+```text
+SOFA FIR
+  -> CanonicalHrtf
+  -> 48 kHz / one radius shell / delay-phase policy
+  -> 64-QMF / 77-hybrid projection
+  -> fifth-order ACN/N3D real-SH field
+  -> per-object direct + early reflections
+  -> shared unitary-FDN late room
+  -> float64 stereo
+```
+
+## Inputs
+
+The CLI has three mutually exclusive HRTF input sources; with none given, a
+default rule resolves the input:
+
+```powershell
+# 1) SOFA: defaults to HRTF/binaural.sofa, or an explicit path
+python main.py input.m4a --binaural
+python main.py input.m4a --binaural --sofa-hrtf C:\HRTF\subject.sofa
+
+# 2) Rosella .personalized_headphone: defaults to HRTF/binaural.personalized_headphone
+python main.py input.m4a --binaural --personalized-headphone
+python main.py input.m4a --binaural --personalized-headphone C:\HRTF\subject.personalized_headphone
+
+# 3) .jochrtf: explicitly load a compiled cache
+python main.py input.m4a --binaural `
+  --compiled-hrtf-cache C:\HRTF\subject.jochrtf
+
+# Optional: create/reuse a transparent disk cache for SOFA
+python main.py input.m4a --binaural --sofa-hrtf C:\HRTF\subject.sofa `
+  --hrtf-cache-policy disk
+```
+
+The default order is `HRTF/binaural.sofa`, then the unique `.jochrtf` under
+`output/hrtf-cache`, then `HRTF/binaural.personalized_headphone`; if none of
+the three exist, an error asks for an explicit path. Multiple `.jochrtf` files
+under `output/hrtf-cache` are also an error requiring an explicit choice.
+
+The `.personalized_headphone` JSON parsing is implemented by this project
+(`src/rosella_model.py`) and does not invoke any Dolby software.
+
+`--hrtf-cache-policy` accepts `none`, `memory`, or `disk`. The default is
+`memory`; neither `none` nor `memory` creates a file. `disk` writes to
+`output/hrtf-cache` by default, or to `--hrtf-cache-dir`. `--hrtf-radius-m`
+selects the nearest measurement-radius shell.
+
+The Python API also uses explicit factories:
+
+```python
+from sofa_binaural_backend import SofaBinauralBackend
+
+renderer = SofaBinauralBackend.from_sofa(
+    "subject.sofa",
+    source_count=16,
+    default_profile="mid",
+    cache_policy="memory",
+)
+
+cached = SofaBinauralBackend.from_compiled_cache(
+    "subject.jochrtf",
+    source_count=16,
+    default_profile="mid",
+)
+```
+
+The factories never guess a format from an unknown suffix: SOFA and `.jochrtf`
+always use distinct loaders.
+
+## Binaural render mode
+
+`--binaural-mode off|near|mid|far` (default `mid`) is a **human-specified
+rendering hint**, not original binaural metadata extracted or recovered from the
+input E-AC-3 JOC bitstream:
+
+- Direct binaural rendering (`--binaural`): near/mid/far apply, default `mid`;
+  `off` is an error;
+- ADM BWF: the low 3 binaural-render-mode bits of the last 15 JOC object entries
+  in DBMD segment 10 carry `off=0/near=1/far=2/mid=3`, leaving the first 10 bed
+  entries unchanged; the default is `mid`, and `off` explicitly disables the
+  binaural metadata hint.
+
+## Canonical SOFA contract
+
+The strict importer currently accepts:
+
+- `Conventions=SOFA`;
+- `SOFAConventions=SimpleFreeFieldHRIR`, version `0.4`, `1.0`, or `1.1`;
+- `DataType=FIR` and `Data.IR[M,2,N]`;
+- one positive finite `Data.SamplingRate` in hertz/Hz;
+- spherical or Cartesian `SourcePosition`;
+- singleton or per-measurement `ListenerPosition/View/Up`;
+- two receivers whose listener-local lateral geometry uniquely identifies L/R;
+- one zero-offset emitter;
+- causal `Data.Delay[I,2]` or `[M,2]`;
+- an explicitly free-field/anechoic `RoomType`.
+
+Receiver order comes from geometry, never from the receiver array index. SOFA
+listener coordinates are $+X$ front, $+Y$ left, $+Z$ up; ADM coordinates are
+$+X$ right, $+Y$ front, $+Z$ up:
+
+$$\bigl(x_{\mathrm{SOFA}},\ y_{\mathrm{SOFA}},\ z_{\mathrm{SOFA}}\bigr) = \bigl(y_{\mathrm{ADM}},\ -x_{\mathrm{ADM}},\ z_{\mathrm{ADM}}\bigr)$$
+
+`CanonicalHrtf` keeps `Data.IR` and `Data.Delay` separate. Only a time-domain
+baseline calls `materialized_measurement()` to apply delay once; the runtime SH
+path never materializes and then restores the delay. Non-48-kHz HRIRs are
+normalized with float64 `scipy.signal.resample_poly`, and delay samples scale by
+the same ratio.
+
+GeneralFIR, BRIR, TF, multiple emitters, ambiguous receivers, and non-free-field
+data require convention-specific adapters. They cannot enter the core importer
+through a reshape.
+
+## Exactly-once delay and phase
+
+The compiler recognizes three mutually exclusive representations:
+
+1. Nonzero `Data.Delay` is external to `Data.IR`; the FIR is not de-rotated and
+   runtime applies the delay once.
+2. With `Data.Delay=0` and an ordinary positive-onset HRIR, each ear's main peak
+   supplies arrival time. Compilation separates it and runtime restores it once.
+   The current threshold is a peak index greater than two samples.
+3. With `Data.Delay=0` and both FIRs at a shared sample-zero origin, no external
+   delay is invented. The authored complex phase stays in the fifth-order field.
+
+No path may add a second ear delay or phase-group delay.
+
+## Public filterbank and directional field
+
+The runtime is fixed at:
+
+- 48 kHz;
+- a 64-sample QMF hop;
+- 64-QMF / 77 hybrid bands;
+- 961 samples of analysis/synthesis latency;
+- fifth order, 36 terms, ACN/N3D real spherical harmonics;
+- float64 PCM, delay, SH, and room state; complex128 band transfers and spectra.
+
+Real and imaginary unit gains for every hybrid band pass through the same
+analysis/synthesis chain to form a 154-real-parameter impulse dictionary. The
+compiler does not sample 77 FFT bins. Defaults are `1e-3` projection ridge and
+`1e-5` SH ridge. Coincident directions are merged before a spherical-Voronoi
+weighted ridge fit.
+
+The fixed resource is `data/rosella_kernels.npz`, which implements publicly
+standardized filter banks, computable from the following formulas.
+
+The hybrid analysis kernels are defined in [3GPP TS 26.405 / ETSI TS 126 405](https://www.etsi.org/deliver/etsi_ts/126400_126499/126405/06.00.00_60/ts_126405v060000p.pdf),
+Section 5.2.2 (Table 1 $Q=8$/$Q=4$ coefficients, delay 6):
+
+$$G_q^p[n] = g^p[n]\cdot\exp\!\Bigl(j\,\frac{2\pi}{Q^p}\bigl(q+\tfrac12\bigr)(n-6)\Bigr),\qquad n=0,\dots,12$$
+
+The QMF analysis table is the MPEG-4 AAC/SBR 64 complex QMF bank of
+ISO/IEC 14496-3/AMD1:2003, subclause 4.B.18.2, stored as the polyphase
+reordering of the public 640-tap prototype $c_0,\dots,c_{639}$:
+
+$$A_{r,t} = \frac{(-1)^t}{128}\,c_{63-r+64t},\qquad r=0,\dots,63,\ t=0,\dots,9$$
+
+The QMF synthesis table is the causal left inverse of the analysis polyphase
+matrix $\mathbf{A}$, i.e. the solution of $\mathbf{A}\,\mathbf{W}=\mathbf{P}$
+($\mathbf{P}$ is the 577-sample delay permutation; total latency
+$961 = 577 + 6\times64$), stored as a rank-4 factorization:
+
+$$W_{b,l} = \sum_{r=1}^{4} t_{b,l,r}\,\mathbf{b}_{b,r}^{\top}$$
+
+The hybrid synthesis table is the 77→64 recombination: identity for the high
+bands, $Y_{3+b}=X_{16+b}$, and for the low bands ($C_p$ is the $8+4+4$ child
+partition):
+
+$$Y_p = \sum_{q\in C_p}\Bigl(\operatorname{Re}X_q + j\,s_q\,\operatorname{Im}X_q\Bigr),\qquad s_q\in\{\pm1\}$$
+
+The loader verifies the archive and every array by SHA-256; the table version,
+all array hashes, and the 77 reference band-center values are part of the cache
+key. Public availability of a standard does not by itself grant permission to
+practice related patent claims. See
+[`data/README.en.md`](../data/README.en.md) and
+[`THIRD_PARTY_NOTICES.md`](../THIRD_PARTY_NOTICES.md) for the sources and the
+rights boundary.
+
+## `.jochrtf`
+
+A `.jochrtf` file is a pickle-free compressed NumPy archive with an exact member set:
+
+| key | dtype / shape |
+|---|---|
+| `metadata_json` | NumPy Unicode scalar containing JSON text (`dtype.kind == "U"`) |
+| `band_center_frequencies_hz` | little-endian `float64[77]` |
+| `coefficients` | little-endian `complex128[36,2,77]` |
+| `delay_coefficients` | little-endian `float64[36,2]` |
+| `delay_bounds` | little-endian `float64[2,2]` |
+
+Metadata uses the `JOC-HRTF-CACHE` magic and records the schema, compiler and
+phase-policy versions, ACN/N3D convention, filterbank hashes, SOFA content
+SHA-256, sample rate, radius, order, both ridge values, payload hash, and fit
+report. Every setting that changes compilation participates in the cache key.
+Metadata never persists an absolute local `source_path`; it may keep a display
+name only.
+
+Before constructing a field, the loader uses `allow_pickle=False` and validates
+ZIP members and expanded sizes, shapes, dtypes, byte order, contiguous layout,
+finite values, delay bounds, band centers, payload hash, and cache key. The
+writer uses a same-directory temporary file, `fsync`, a process-held OS file
+lock, and atomic `os.replace`. Its hidden `.lock` sidecar may remain and does not
+mean that a writer still owns the lock. Outdated, damaged, or mismatched
+caches cannot hit. SOFA input rebuilds an invalid cache; an explicitly selected
+cache reports the error.
+
+Deleting a disk cache must not change the field or render produced from the same
+SOFA and compiler configuration.
+
+A `.jochrtf` file contains directional-field coefficients and delay data
+transformed from the source HRIRs. Its reproducibility therefore does not make
+it licence-free. Creating a cache does not enlarge the rights granted by the
+source SOFA/HRTF dataset: use, copying, and redistribution remain subject to
+that dataset's terms. If those terms are unclear, keep `.jochrtf` as a private
+local cache and do not ship it with the program or another build artifact.
+`source_sha256` is only a content-integrity identifier, not proof of provenance
+or permission.
+
+## JOC objects and room behavior
+
+The production adapter retains the existing JOC schedule:
+
+- `[1536,16]` input per frame;
+- channel 0 is special LFE and channels 1..15 are JOC objects;
+- ID11/OAMD positions use a sample-timed timeline;
+- source parameters update every 512 samples;
+- every object owns independent direct/early history while one late FDN is shared;
+- `finish()` drains early/late tails; output gain is explicit, with no implicit
+  limiter or programme loudness normalization.
+
+Near/Mid/Far, equal-power direct level, six first-order shoebox image sources,
+late sends, the unitary FDN, the 120–180 Hz cosine-squared LFE low-pass, and room
+calibration are JOC project-defined behavior, not constants published by SOFA or
+Dolby.
+
+The public SOFA binaural renderer defaults to the C++20 native core under
+`--backend auto/native` (`ejoc_sofa_binaural_*` in `lib/eac3joc_core.dll`): the
+filterbank, the SH direction-field evaluation, the per-object direct/early
+histories and the shared FDN all run natively, while Python only compiles the
+SOFA source and issues the per-512-sample metadata updates. When the native
+library is unavailable the renderer falls back to the Python/NumPy reference
+implementation; the two agree to better than 1e-9. `--backend python` forces
+the Python backend.
+`--backend` still selects native/Python JOC reconstruction and speaker rendering;
+native acceleration for the public binaural DSP is outside the current API.
+
+## Technical references and rights boundary
+
+- [SOFA SimpleFreeFieldHRIR convention](https://www.sofaconventions.org/mediawiki/index.php/SimpleFreeFieldHRIR)
+- [3GPP TS 26.405 / ETSI TS 126 405 (64-QMF/77-hybrid definition)](https://www.etsi.org/deliver/etsi_ts/126400_126499/126405/06.00.00_60/ts_126405v060000p.pdf)
+- [Dolby binaural render-mode workflow](https://professionalsupport.dolby.com/s/article/What-is-Binaural-Render-Mode-and-how-do-the-settings-affect-my-mix)
+- [EP3090576A1](https://patents.google.com/patent/EP3090576A1/en), used only as
+  architectural background for direct/early/late, subbands, and FDNs; it does
+  not establish that any product uses a particular embodiment.
+
+Public availability of a specification, source file, or patent document does
+not by itself authorize copying its contents, redistribution of derivatives,
+or practice of patent claims. These technical references grant no patent
+licence and make no non-infringement representation. Anyone preparing a release
+or product integration must assess the applicable data and software licences,
+patent permissions, and freedom to operate. See
+[`THIRD_PARTY_NOTICES.md`](../THIRD_PARTY_NOTICES.md) for the public-standard
+provenance and rights boundary.
@@ -0,0 +1,238 @@
+# 双耳渲染
+
+[English](binaural.en.md) · [返回 README](../README.md)
+
+JustOneCacophony 的双耳后端支持三种 HRTF 来源：`SimpleFreeFieldHRIR` SOFA、
+杜比官方软件个性化扫描导出的 Rosella `.personalized_headphone`（JSON 解析由本项目
+自行实现，不调用杜比软件），以及从 SOFA 编译出的 `.jochrtf` 缓存。SOFA 在模型加载
+时编译成内存方向场；`.jochrtf` 只是可删除、可重建的 JOC compiled HRTF cache，
+不是交换格式，也不是使用 SOFA 的前置步骤。
+
+```text
+SOFA FIR
+  -> CanonicalHrtf
+  -> 48 kHz / 单 radius shell / delay-phase policy
+  -> 64-QMF / 77-hybrid projection
+  -> 五阶 ACN/N3D 实球谐场
+  -> 逐对象 direct + early reflections
+  -> shared unitary-FDN late room
+  -> stereo float64
+```
+
+## 输入接口
+
+CLI 有三个互斥的 HRTF 输入来源；都不指定时按默认规则自动选择：
+
+```powershell
+# 1) SOFA：缺省取 HRTF/binaural.sofa，也可显式指定
+python main.py input.m4a --binaural
+python main.py input.m4a --binaural --sofa-hrtf C:\HRTF\subject.sofa
+
+# 2) Rosella .personalized_headphone：缺省取 HRTF/binaural.personalized_headphone
+python main.py input.m4a --binaural --personalized-headphone
+python main.py input.m4a --binaural --personalized-headphone C:\HRTF\subject.personalized_headphone
+
+# 3) .jochrtf：显式读取预编译 cache
+python main.py input.m4a --binaural `
+  --compiled-hrtf-cache C:\HRTF\subject.jochrtf
+
+# 可选：SOFA 透明生成/复用磁盘 cache
+python main.py input.m4a --binaural --sofa-hrtf C:\HRTF\subject.sofa `
+  --hrtf-cache-policy disk
+```
+
+默认选择顺序：`HRTF/binaural.sofa` → `output/hrtf-cache` 下唯一的 `.jochrtf` →
+`HRTF/binaural.personalized_headphone`；三者都没有时报错并提示显式指定。
+`output/hrtf-cache` 下有多个 `.jochrtf` 时同样报错，要求显式选择。
+
+`.personalized_headphone` 的 JSON 解析由本项目自行实现（`src/rosella_model.py`），
+不调用任何杜比软件。
+
+`--hrtf-cache-policy` 可取 `none`、`memory`、`disk`。默认是 `memory`；`none` 和
+`memory` 都不会创建磁盘文件。`disk` 默认写入 `output/hrtf-cache`，也可用
+`--hrtf-cache-dir` 指定。`--hrtf-radius-m` 选择距离目标最近的 measurement shell。
+
+Python API 使用显式 factory：
+
+```python
+from sofa_binaural_backend import SofaBinauralBackend
+
+renderer = SofaBinauralBackend.from_sofa(
+    "subject.sofa",
+    source_count=16,
+    default_profile="mid",
+    cache_policy="memory",
+)
+
+cached = SofaBinauralBackend.from_compiled_cache(
+    "subject.jochrtf",
+    source_count=16,
+    default_profile="mid",
+)
+```
+
+文件工厂不会按“未知后缀”猜格式：SOFA 和 `.jochrtf` 始终走不同 loader。
+
+## 双耳渲染模式
+
+`--binaural-mode off|near|mid|far`（默认 `mid`）是**人为指定的渲染提示**，不是
+从输入 E-AC-3 JOC 码流提取或还原的原始双耳元数据：
+
+- 直接双耳渲染（`--binaural`）：near/mid/far 生效，默认 `mid`；`off` 报错；
+- ADM BWF：DBMD segment 10 中后 15 个 JOC 对象的 binaural render mode 写
+  `off=0/near=1/far=2/mid=3`，前 10 个 bed 保持不变，默认 `mid`；`off` 用于显式
+  关闭双耳元数据提示。
+
+## Canonical SOFA 契约
+
+当前 strict importer 接受：
+
+- `Conventions=SOFA`；
+- `SOFAConventions=SimpleFreeFieldHRIR`，version `0.4`、`1.0` 或 `1.1`；
+- `DataType=FIR`，`Data.IR[M,2,N]`；
+- 单一正有限 `Data.SamplingRate`，单位为 hertz/Hz；
+- spherical 或 Cartesian `SourcePosition`；
+- 单值或 per-measurement 的 `ListenerPosition/View/Up`；
+- 两个能由 listener-local lateral 坐标唯一识别左右的 receiver；
+- 单一且零偏移的 emitter；
+- causal `Data.Delay[I,2]` 或 `[M,2]`；
+- 明确的 free-field/anechoic `RoomType`。
+
+receiver 左右顺序由几何决定，不能假定 `Data.IR` 的 receiver index。SOFA listener
+坐标为 $+X$ front、$+Y$ left、$+Z$ up；ADM 坐标为 $+X$ right、$+Y$ front、
+$+Z$ up，转换为：
+
+$$\bigl(x_{\mathrm{SOFA}},\ y_{\mathrm{SOFA}},\ z_{\mathrm{SOFA}}\bigr) = \bigl(y_{\mathrm{ADM}},\ -x_{\mathrm{ADM}},\ z_{\mathrm{ADM}}\bigr)$$
+
+`CanonicalHrtf` 将 `Data.IR` 与 `Data.Delay` 分开保存。只有时域 baseline 才调用
+`materialized_measurement()` 将 delay 应用一次；运行时 SH 路径不先 materialize。
+非 48 kHz HRIR 使用 float64 `scipy.signal.resample_poly` 规范化，delay samples 按
+相同比例缩放。
+
+GeneralFIR、BRIR、TF、多 emitter、多义 receiver 或非 free-field 数据需要单独的
+convention adapter，不能只通过 reshape 进入核心 importer。
+
+## Delay/phase：exactly once
+
+编译器只允许三种互斥语义：
+
+1. 非零 `Data.Delay` 是 `Data.IR` 外部 delay；FIR 不去旋，运行时应用一次。
+2. `Data.Delay=0` 且 HRIR 有普通正 onset：以每耳 main peak 分离 arrival，拟合后
+   在运行时恢复一次；当前阈值为 peak index 大于 2 samples。
+3. `Data.Delay=0` 且双耳 FIR 共享 sample-0 起点：不发明外部 delay，原 complex
+   phase 直接进入五阶场。
+
+任何路径都不能再叠加第二套 ear delay 或 phase-group delay。
+
+## 公开 filterbank 与方向场
+
+运行时固定为：
+
+- 48 kHz；
+- 64-sample QMF hop；
+- 64-QMF / 77-hybrid；
+- analysis/synthesis latency 961 samples；
+- 五阶、36 项、ACN/N3D real spherical harmonics；
+- PCM、delay、SH、room state 为 float64；频带传递和频域状态为 complex128。
+
+每个 hybrid band 的 real/imaginary 单位增益都通过同一套 analysis/synthesis 链生成
+脉冲字典，共 154 个实参数；编译不是直接读取 77 个 FFT bin。默认 projection
+ridge 为 `1e-3`，SH ridge 为 `1e-5`。同方向 measurement 先合并，再用球面 Voronoi
+面积权重做 ridge fit。
+
+固定表位于 `data/rosella_kernels.npz`，实现公开标准化的滤波器组，各表可由如下
+公式计算。
+
+hybrid 分析核定义于 [3GPP TS 26.405 / ETSI TS 126 405](https://www.etsi.org/deliver/etsi_ts/126400_126499/126405/06.00.00_60/ts_126405v060000p.pdf)
+第 5.2.2 节（Table 1 的 $Q=8$/$Q=4$ 系数，delay 6）：
+
+$$G_q^p[n] = g^p[n]\cdot\exp\!\Bigl(j\,\frac{2\pi}{Q^p}\bigl(q+\tfrac12\bigr)(n-6)\Bigr),\qquad n=0,\dots,12$$
+
+QMF analysis 表即 MPEG-4 AAC/SBR（ISO/IEC 14496-3/AMD1:2003 第 4.B.18.2 节）
+的 64 complex QMF bank；打包的 $64\times10$ 表是公开 640-tap prototype
+$c_0,\dots,c_{639}$ 的多相重排：
+
+$$A_{r,t} = \frac{(-1)^t}{128}\,c_{63-r+64t},\qquad r=0,\dots,63,\ t=0,\dots,9$$
+
+QMF synthesis 表为上述 analysis 多相矩阵 $\mathbf{A}$ 的因果左逆，即求解
+$\mathbf{A}\,\mathbf{W}=\mathbf{P}$（$\mathbf{P}$ 为 577-sample 延迟置换；
+全链 $961 = 577 + 6\times64$），以 rank-4 分解形式存储：
+
+$$W_{b,l} = \sum_{r=1}^{4} t_{b,l,r}\,\mathbf{b}_{b,r}^{\top}$$
+
+hybrid synthesis 表为 77→64 重组：高频带恒等 $Y_{3+b}=X_{16+b}$；低频带
+（$C_p$ 为 $8+4+4$ 子带划分）：
+
+$$Y_p = \sum_{q\in C_p}\Bigl(\operatorname{Re}X_q + j\,s_q\,\operatorname{Im}X_q\Bigr),\qquad s_q\in\{\pm1\}$$
+
+loader 校验 archive 和每个数组的 SHA-256；table version、所有数组 hash 与
+77 个 band-center 参考值都属于 cache key。标准可公开获取不等于获准实施相关
+专利；更多来源信息见 [`data/README.md`](../data/README.md) 与
+[`THIRD_PARTY_NOTICES.md`](../THIRD_PARTY_NOTICES.md)。
+
+## `.jochrtf`
+
+`.jochrtf` 是无 pickle 的压缩 NumPy archive，固定包含：
+
+| key | dtype / shape |
+|---|---|
+| `metadata_json` | 含 JSON 文本的 NumPy Unicode scalar（`dtype.kind == "U"`） |
+| `band_center_frequencies_hz` | little-endian `float64[77]` |
+| `coefficients` | little-endian `complex128[36,2,77]` |
+| `delay_coefficients` | little-endian `float64[36,2]` |
+| `delay_bounds` | little-endian `float64[2,2]` |
+
+metadata magic 固定为 `JOC-HRTF-CACHE`，并记录 schema/compiler/phase-policy、
+ACN/N3D、filterbank table hashes、SOFA content SHA-256、采样率、radius、order、
+两个 ridge、payload hash 和 fit report。cache key 覆盖所有会改变编译结果的字段。
+metadata 不保存本机绝对 `source_path`，仅可保存 source display name。
+
+loader 使用 `allow_pickle=False`，并在构造对象前检查 ZIP 成员集、解压大小、shape、
+dtype、端序、连续布局、有限值、delay bounds、band centers、payload hash 和 cache
+key。writer 使用同目录临时文件、`fsync`、进程持有的 OS 文件锁和原子
+`os.replace`；对应的隐藏 `.lock` sidecar 可保留，但不代表仍有 writer 持锁。
+旧版本、损坏或配置不匹配的 cache 不能命中；从 SOFA 启动时会重建，显式 cache
+入口则直接报错。
+
+删除磁盘 cache 后，从同一 SOFA 和同一编译配置得到的场与渲染结果不得改变。
+
+`.jochrtf` 包含由源 HRIR 变换得到的方向场系数与 delay 数据，因此“可以重建”不表示
+它不受数据许可约束。生成 cache 不会扩大源 SOFA/HRTF 数据集授予的权利；cache 的
+使用、复制和再分发仍须遵守源数据集条款。不能确认条款时，应把 `.jochrtf` 作为本地
+私有 cache，不随程序或构建产物发布。`source_sha256` 只用于内容一致性校验，不是许可
+或来源证明。
+
+## JOC 对象与房间
+
+生产适配器继续使用现有 JOC 调度：
+
+- 每帧输入 `[1536,16]`；
+- channel 0 是 special LFE，channel 1..15 是 JOC objects；
+- ID11/OAMD position 使用 sample-timed timeline；
+- 每 512 samples 更新方向/profile；
+- 每个对象拥有独立 direct/early history，late FDN 全局共享；
+- `finish()` 排空 early/late tail；输出增益显式应用，不隐含 limiter 或节目响度归一化。
+
+Near/Mid/Far、equal-power direct level、六面 shoebox 一阶 image source、late send、
+unitary FDN、LFE 120–180 Hz cosine-squared 低通及 room calibration 都是 JOC
+项目定义行为，不是 SOFA 或 Dolby 公布常数。
+
+公开 SOFA 双耳渲染在 `--backend auto/native` 下默认走 C++20 原生核
+（`lib/eac3joc_core.dll` 的 `ejoc_sofa_binaural_*` 接口：filterbank、SH 方向场求值、
+逐对象 early/direct 历史与共享 FDN 全部在原生侧执行，Python 只做 SOFA 编译与每
+512-sample 的元数据更新）；原生库不可用时自动回退 Python/NumPy 参考实现，两者
+逐值一致（差异 < 1e-9）。`--backend python` 强制使用 Python 后端。
+
+## 技术引用与权利边界
+
+- [SOFA SimpleFreeFieldHRIR convention](https://www.sofaconventions.org/mediawiki/index.php/SimpleFreeFieldHRIR)
+- [3GPP TS 26.405 / ETSI TS 126 405（64-QMF/77-hybrid 定义）](https://www.etsi.org/deliver/etsi_ts/126400_126499/126405/06.00.00_60/ts_126405v060000p.pdf)
+- [Dolby binaural render mode workflow](https://professionalsupport.dolby.com/s/article/What-is-Binaural-Render-Mode-and-how-do-the-settings-affect-my-mix)
+- [EP3090576A1](https://patents.google.com/patent/EP3090576A1/en)，仅作 direct/early/late、
+  subband 与 FDN 架构背景，不证明某个产品使用特定实施例。
+
+规范、源码或专利文献可公开获取，不等于获准复制其内容、再分发派生产物或实施其中的
+专利权利要求。本项目的技术引用本身不授予专利许可，也不作不侵权保证；准备发布或集成
+到产品的一方应自行审查适用的数据许可、软件许可、专利许可及 freedom-to-operate。
+公开标准来源与权利边界见
+[`THIRD_PARTY_NOTICES.md`](../THIRD_PARTY_NOTICES.md)。
@@ -116,7 +116,26 @@ Supported layouts:
 2.0  3.1  5.1  7.1  5.1.2  5.1.4  7.1.2  7.1.4  9.1.4  9.1.6
 ```

-## 6. Building
+## 6. Binaural-rendering ABI
+
+The shared library provides a 512-sample float64 binaural DSP interface:
+
+```c
+ejoc_binaural_renderer_handle ejoc_binaural_renderer_create(void);
+int ejoc_binaural_renderer_configure_kernels(...);
+int ejoc_binaural_renderer_configure_room(...);
+int ejoc_binaural_renderer_process(
+    ejoc_binaural_renderer_handle handle,
+    const double* input16_interleaved,   /* [512][16] */
+    const double* gains_complex,         /* [16][2][77][2] */
+    const double* room_sends,            /* [16] */
+    double output_gain,
+    double* output_stereo_interleaved);  /* [512][2] */
+```
+
+Python parses the model, evaluates the OAMD timeline, and supplies complex gains and room sends every 512 samples. The C++ handle owns QMF, hybrid, recursive-room, and QMF-synthesis state. Inputs, state, accumulation, and output are double/complex double.
+
+## 7. Building

 The CMake definition is `native/CMakeLists.txt`. Run from the repository root:

@@ -138,7 +157,7 @@ The MSVC configuration uses the static CRT. Other runtime dependencies depend on

 The repository does not include native binaries by default. A prebuilt Release runtime or a locally built runtime can be placed directly under `lib/`.

-## 7. Runtime lookup and fallback
+## 8. Runtime lookup and fallback

 Lookup order:

@@ -148,7 +167,7 @@ Lookup order:

 `--backend auto` falls back to NumPy when loading fails, and `--backend python` skips native discovery. The current CLI also prints the failure and falls back for `--backend native`; this existing behavior should not be read as successful native execution.

-## 8. Implementation boundaries
+## 9. Implementation boundaries

 - The native layer accepts only dense-JOC data already parsed by Python.
 - The ABI fixes a 1536-sample JOC frame, at most 15 objects, at most 23 parameter bands, and at most 2 data points.
@@ -116,7 +116,39 @@ int ejoc_speaker_renderer_process(
 2.0  3.1  5.1  7.1  5.1.2  5.1.4  7.1.2  7.1.4  9.1.4  9.1.6
 ```

-## 6. 构建
+## 6. 双耳渲染 ABI
+
+共享库提供 512-sample float64 双耳 DSP：
+
+```c
+ejoc_binaural_renderer_handle ejoc_binaural_renderer_create(void);
+int ejoc_binaural_renderer_configure_kernels(...);
+int ejoc_binaural_renderer_configure_room(...);
+int ejoc_binaural_renderer_process(
+    ejoc_binaural_renderer_handle handle,
+    const double* input16_interleaved,   /* [512][16] */
+    const double* gains_complex,         /* [16][2][77][2] */
+    const double* room_sends,            /* [16] */
+    double output_gain,
+    double* output_stereo_interleaved);  /* [512][2] */
+```
+
+Python 负责模型解析、OAMD 时间轴和每 512 samples 的 complex gains/room sends。C++ handle 保存 QMF、hybrid、递归 room 和 QMF synthesis 状态。全部输入、状态、乘加和输出均为 double/complex double。
+
+## 6.1 公开 SOFA 双耳渲染 ABI
+
+共享库同时提供完整的原生 SOFA 双耳渲染器（`ejoc_sofa_binaural_*`），它镜像
+Python `SofaBinauralBackend` 的全部数学：64-QMF/77-hybrid analysis/synthesis、
+五阶 ACN/N3D 实球谐方向场求值、whole-QMF-slot 逐对象 delay 历史、六面一阶
+image-source early reflections、共享 unitary FDN late room、LFE 120–180 Hz
+低通与 961-sample latency 语义。kernel 表、编译好的 HRTF 场与房间常数通过
+`configure_kernels/configure_field/configure_room` 一次上传；每 512-sample
+block 先 `set_source` 更新 16 个 source，再 `process` 输入 PCM；`process` 返回
+裁剪后的 stereo 样本数（首个 961 samples 被丢弃）。`finish` 以 64-sample 对齐的
+块排空尾音。Python 桥位于 `src/sofa_native_backend.py`，与 Python 参考实现逐值
+一致（差异 < 1e-9）；原生库缺失时 `main.py` 自动回退 Python。
+
+## 7. 构建

 CMake 定义位于 `native/CMakeLists.txt`。从仓库根目录运行：

@@ -138,7 +170,7 @@ MSVC 配置使用静态 CRT。其他运行时依赖由平台和工具链决定

 仓库默认不附带原生二进制。预构建的 Release 运行库或自行构建的运行库均可直接放入 `lib/`。

-## 7. 运行时查找与回退
+## 8. 运行时查找与回退

 查找顺序为：

@@ -148,7 +180,7 @@ MSVC 配置使用静态 CRT。其他运行时依赖由平台和工具链决定

 `--backend auto` 在加载失败时回退到 NumPy；`--backend python` 跳过原生探测。`--backend native` 当前也会打印失败原因后回退，这是现有 CLI 行为，不应理解为原生库已成功使用。

-## 8. 实现边界
+## 9. 实现边界

 - 原生层只接收 Python 已解析的 dense JOC 数据。
 - ABI 固定了 1536-sample JOC 帧、最多 15 个对象、最多 23 个参数带和最多 2 个数据点。