Poetry3D – Transforms poems into 3D semantic trees

Summary: Poetry3D converts poems into three-dimensional semantic trees by mapping word embeddings into 3D space, visualizing the relationships between words based on their meanings. It uses AI concepts like tokenization, vector embeddings, and cosine similarity to create unique visual structures representing each poem.

What it does

Poetry3D processes poems by treating each word as a token, embedding them in a high-dimensional space, then projecting these embeddings into 3D to form tree-like structures where semantically similar words cluster as branches. Users can export these visualizations as PNG images or OBJ files for 3D printing or digital art.

Who it's for

This tool is designed for those interested in exploring the semantic structure of poetry through AI-driven visualization, including artists, poets, and AI enthusiasts.

Why it matters

It reveals the hidden geometric relationships in language, bridging AI data processing with human interpretation of poetic meaning.