Sonification of IC 1231: A Multimodal Dataset for AI-Mediated Analysis of Astronomical Images

This dataset presents a complete multimodal research artifact demonstrating how non-linguistic visual data can be made analyzable by language models through rule-based data sonification and explicit textual documentation. The core artifact is a raw audio signal generated via deterministic sonification of the astronomical image IC 1231, using declared mappings between image features and audio parameters (spatial position → time and frequency, luminance → amplitude, morphology → spectral texture). No musical quantization, harmonic tuning, rhythmic grid, or aesthetic constraints were applied. The dataset includes complementary materials: a formal signal-analysis report, a textual semantic layer designed for language-model reasoning, a slide deck, an explanatory infographic, and a narrated video. Together, these components form a closed interpretive system in which the audio functions as a non-self-describing signal and the documentation provides the necessary symbolic interface for grounded analysis. Quantitative time–spectral analysis confirms non-trivial correlations between scan progression and spectral features, demonstrating preservation of image topology in the audio domain. The dataset is intended as a methodological reference for AI-mediated analysis of raw signals, not as a musical or artistic work.