# Rights and credits

Project explanations, graphics and helper scripts: Carsten Püttbach, 2026.
Provided as the accompanying learning material for the published episode.
No new blanket Creative Commons licence is declared by this release.
Original OpenFOAM notices and upstream terms remain applicable to its files;
the supplied licence text is retained. Cited papers and upstream software are
not relicensed by this download. No third-party paper PDFs, model weights or
software binaries are included. Any further redistribution must preserve the
applicable notices and terms.

## Production provenance

# Production provenance and credits

Actual CFD: installed OpenCFD v2606 icoFoam; original case inputs and logs retained. Graphics: project code with Python, NumPy, Matplotlib and Pillow. Encoding: FFmpeg. Workbook: ReportLab, with embedded DejaVu fonts. No 3Blue1Brown or Veritasium footage, logos, music, characters or font files are reused.

Narration: Puck (am_puck), Kokoro v1.0 model, local kokoro-onnx 0.6.1, speed 1.00. The official upstream model card labels the weights Apache-2.0: https://huggingface.co/hexgrad/Kokoro-82M . The local inference project lists MIT: https://github.com/thewh1teagle/kokoro-onnx . Model hashes and generated audio hashes are retained. These observations describe upstream metadata, not a blanket legal opinion about every downstream use.

Music: original project-specific procedural score, 64 BPM, no external audio samples. The composition source, deterministic seed and audio hash are recorded in music-provenance.json. No paid TTS or music service was invoked.

The learning bundle contains configuration and project scripts. Original OpenFOAM file headers are preserved. It does not redistribute the OpenFOAM application, TTS weights or third-party paper PDFs. Source and licence obligations should remain attached to any future distributed software package.