[aguirreSL]

Summary

Adds four example notebooks in examples/notebooks/, run with their outputs, for the new functions of PyTTa. Each one compares its results with a known value, so a reader can see what to expect and check an installation:

1. 01_simulated_measurement.ipynb: an impulse response measured with an exponential sweep through SimulatedFRFMeasure, without a sound card, from a synthetic room with T = 1.2 s. The chain adds a cubic nonlinearity, noise 70 dB down and 24-bit quantization. The measured T30 is 1.14 to 1.22 s in the octave bands from 125 Hz to 4 kHz.
2. 02_room_acoustics.ipynb: the same 6 × 5 × 4 m room with α = 0.2 by image sources, by ray tracing with diffuse walls and by the statistical model, against Sabine (0.653 s) and Eyring (0.585 s). Ray tracing gives T30 = 0.611 s, the statistical model 0.57 to 0.61 s, and the image sources, with specular walls, a slower decay, as expected for a shoebox.
3. 03_psychoacoustics.ipynb: the reference sounds of the psychoacoustic units. The 1 kHz tone at 40 dB gives 1.000 sone (40.00 phon), the critical band noise at 1 kHz and 60 dB 1.020 acum, the tone modulated at 70 Hz 1.000 asper, and the tone modulated at 4 Hz 0.941 vacil.
4. 04_fem_bem_sphere.ipynb: a DistMesh mesh of the unit ball for both solvers. The FEM gives the first nine modes of the rigid sphere within 0.7 to 2.3 % of the roots of j_l'(ka) = 0, and the BEM on the boundary of the same mesh gives the field of a pulsating sphere within 0.6 %.

Dependencies

The notebooks use functions from open pull requests, and they were run on a branch that merges them: #52 (generation on Python 3.12), #79 and #92 (psychoacoustics), #101 (image sources), #109 (statistical impulse response), #117 (simulated measurement), #122 (FEM), #123 (BEM), #124 (ray tracing), #125 (meshes) and #127 (C80 and D50). This pull request can wait for them; it adds no code.

Co-Authored-By: Claude Fable 5.1 noreply@anthropic.com