These interactive tools allow you to explore how time-delay reconstruction works and how geometric patterns can be built from sound recordings. The main demo focuses on bird sounds, while the additional demos provide simple introductions to attractors and embedding.
Recommended starting point: Begin with the sound demo if you want to see how the method is applied to recordings. The Lorenz and Takens demos provide background intuition for the ideas used in the analysis.
Explore how geometric patterns can be reconstructed from sound recordings. This demo allows you to upload an audio file, adjust reconstruction parameters, and visualize the resulting trajectory in 3D.
You can use this demo to:
reconstruct patterns from sound recordings
adjust the embedding dimension and time lag
visualize the reconstructed structure in 3D
compare how parameter changes affect the geometry
This demo introduces the Lorenz system, a classic example in nonlinear dynamics. By varying the model parameters, you can explore how different attractor shapes emerge and how trajectories evolve in phase space.
This demo illustrates the basic idea of time-delay reconstruction. Starting from a single observed signal, delayed copies are used to reconstruct a geometric trajectory, showing how temporal structure can become visible in an embedding space.