Visualization and optimization tools for recording studio acoustic simulations created with go-recording-studio.

This toolkit provides two main functionalities for acoustic studio design:

1. 3D visualization of acoustic simulation results, helping studio designers:

   • View reflection paths and their interaction with room surfaces
   • Identify problematic reflection points
   • Visualize the Reflection-Free Zone (RFZ)
   • Analyze reflection patterns with reduced visual noise

2. Bayesian optimization of studio parameters, enabling:

   • Automated search for optimal speaker and listening positions
   • Multi-process parameter optimization
   • Customizable optimization objectives
   • Constraint-based parameter exploration

The visualization tool processes simulation results from go-recording-studio and provides interactive 3D views using trimesh:

1. Load simulation results from a go-recording-studio output directory
2. Render the room geometry, reflection paths, and RFZ
3. Support interactive viewing with zoom, pan, and rotate capabilities
4. Provide specialized viewing modes:
   • Step mode (--step): View individual reflection paths sequentially
   • Culling mode (--cull): Group similar reflections for clearer visualization

The optimizer uses Bayesian optimization to find optimal studio configurations:

1. Define parameter bounds and constraints
2. Run parallel simulations across multiple processes
3. Primary optimization goals:
   • Achieve minimum 30ms Initial Time Delay (ITD)
   • Minimize required absorptive surface area
4. Support for custom optimization objectives

• You'll have to do some hacking. The optimization code is functional for fellow hackers but really not polished enough for non-hackers.

1. Create and activate a Python virtual environment:

python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

2. Install dependencies:

pip install -r requirements.txt

• Python 3.x
• A completed simulation from go-recording-studio

Basic usage:

python main.py /path/to/experiment [options]

Available options:

--step         Step through reflections one at a time
--cull FLOAT   Cull very similar paths from the render
               (Values between 0.05 and 0.3 meters recommended)
--points       Show the location of the final reflection in each arrival

For each reflection path, the visualization displays:

• ITD (Initial Time Delay) in milliseconds
• Path gain in dB
• Shot gain in dB
• Number of reflections
• Surface name of last reflection

In step mode, navigation controls are:

• 'n': next reflection
• 'p': previous reflection
• 'q': quit

The visualization tool reads three key files from the simulation output:

• annotations.json: Contains reflection paths and geometry data (see annotations_schema.md)
• summary.json: Contains acoustic parameters and results (see summary_schema.md)
• room.stl: Contains acoustic parameters and results (see summary_schema.md)

1. Create a configuration file based on the example config

2. Run the optimizer:

python acoustic_optimizer.py config.yaml

MIT license. (See LICENSE file)

• Uses go-recording-studio for acoustic simulation
• Built with trimesh for 3D visualization
• Optimization powered by scikit-optimize