feat(audio, tools): Add Task #72 and enhance Blender exporter

- Add Task #72 (Audio Pipeline Streamlining) to TODO.md and PROJECT_CONTEXT.md.
- Update blender_export.py to support 'EMPTY' objects for planes and export 'plane_distance'.

3 files changed, 35 insertions, 1 deletions

diff --git a/PROJECT_CONTEXT.md b/PROJECT_CONTEXT.md
index 9b36223..663d320 100644
--- a/PROJECT_CONTEXT.md
+++ b/PROJECT_CONTEXT.md
@@ -50,6 +50,11 @@ Style:
     - Phase 3: File I/O (load .wav/.spec, export procedural_params.txt + C++ code)
     - See `doc/SPECTRAL_BRUSH_EDITOR.md` for complete design
 
+- **Task #72: Audio Pipeline Streamlining**
+    - Optimize data flow: reduce copies and temp buffers.
+    - Direct additive mixing: have tracker add samples directly to output buffer.
+    - Precision: maintain float32 internally, clip/convert at the very end.
+
 - **Visuals & Content**
     - [ ] **Task #52: Procedural SDF Font**: Minimal bezier/spline set for [A-Z, 0-9] and SDF rendering.
     - [ ] **Task #53: Particles Shader Polish**: Improve visual quality of particles.
diff --git a/TODO.md b/TODO.md
index 2400113..72d1b74 100644
--- a/TODO.md
+++ b/TODO.md
@@ -93,6 +93,22 @@ This file tracks prioritized tasks with detailed attack plans.
 
 ---
 
+## Priority 2: Audio Pipeline Streamlining (Task #72)
+
+**Goal**: Optimize the audio pipeline to reduce memory copies and simplify the data flow by using direct additive mixing and deferred clipping.
+
+- [ ] **Phase 1: Direct Additive Mixing**
+  - Modify `Synth` and `Tracker` to accept a target output buffer for direct additive mixing instead of returning isolated voice samples.
+  - Eliminate temporary buffers used for individual voice rendering.
+- [ ] **Phase 2: Float32 Internal Pipeline**
+  - Ensure the entire internal pipeline (synthesis, mixing) maintains full `float32` precision without intermediate clipping.
+- [ ] **Phase 3: Final Clipping & Conversion**
+  - Implement a single, final stage that performs clipping (limiter/clamping) and conversion to `int16` (or other hardware-native formats) just before the audio backend delivery.
+- [ ] **Phase 4: Verification**
+  - Verify audio quality and performance improvements with `test_demo` and existing audio tests.
+
+---
+
 ## Priority 2: 3D System Enhancements (Task #18)
 **Goal:** Establish a pipeline for importing complex 3D scenes to replace hardcoded geometry.
 
diff --git a/tools/blender_export.py b/tools/blender_export.py
index da7b986..c4a45ff 100644
--- a/tools/blender_export.py
+++ b/tools/blender_export.py
@@ -16,6 +16,7 @@ import os
 #   quat rotation (x, y, z, w)
 #   vec3 scale
 #   vec4 color
+#   float plane_distance (if type == PLANE)
 #   uint32_t mesh_name_len
 #   char[] mesh_name (if type == MESH)
 #   float mass
@@ -25,7 +26,13 @@ import os
 def export_scene(filepath):
     print(f"Exporting scene to {filepath}...")
     
-    objects = [obj for obj in bpy.context.scene.objects if obj.visible_get() and obj.type == 'MESH']
+    objects = []
+    for obj in bpy.context.scene.objects:
+        if obj.visible_get():
+            if obj.type == 'MESH':
+                objects.append(obj)
+            elif obj.type == 'EMPTY' and obj.name.lower().startswith('plane_'):
+                objects.append(obj)
     
     with open(filepath, 'wb') as f:
         # Header
@@ -84,6 +91,12 @@ def export_scene(filepath):
                 c = obj.active_material.diffuse_color
                 color = (c[0], c[1], c[2], c[3])
             f.write(struct.pack('<4f', *color))
+
+            # Plane Distance (if type == PLANE)
+            plane_distance = 0.0
+            if obj_type == 2: # PLANE
+                plane_distance = obj.get('plane_distance', 0.0)
+            f.write(struct.pack('<f', plane_distance))
             
             # Mesh Asset Name
             mesh_asset_name = ""