Unbound
Unbound is a music genre discovery app that breaks listeners out of their Spotify suggestion bubbles by mathematically finding the music they are wired to love but have never discovered.
Core trick
Computes cosine similarity between a user's Last.fm tag vector and 40 genre centroids, then KNN picks five unexplored genres for Gemini to explain.
Tags
Description
Unbound is a music genre discovery app that breaks listeners out of their Spotify suggestion bubbles by mathematically finding the music they are wired to love but have never discovered. Instead of relying on AI to hallucinate recommendations, Unbound fetches real listener tags from Last.fm, maps the user's "taste fingerprint," and uses a K-Nearest Neighbors (KNN) algorithm with cosine similarity against 40 global genre centroids to pick the top 5 unexplored matches. Gemini 2.0 Flash is then used purely for the explanation layer, translating the mathematical proximity into beautifully written, emotionally resonant reasons for why the genre fits the user's sound identity. It is the personalized, fully computed successor to the famous app - Every Noise at Once.
README excerpt
# Unbound **[Vercel App](https://unbound-pied.vercel.app)** > "We compute your taste fingerprint from real listener data, run KNN against 40 live genre centroids, and surface the 5 genres your brain is already wired for but has never found." Unbound is a music genre discovery app that breaks you out of your Spotify algorithm bubble. Instead of giving you more of what you already know, Unbound finds the genres you've never explored but are mathematically destined to love. <br /> ## 🎵 How It Works **Gemini Explains. The Math Picks.** 1. **Taste Fingerprint:** You enter your favorite artists. We fetch real listener dataset tags from Last.fm (millions of actual scrobbling behaviors). 2. **Compute (KNN):** We build your taste fingerprint and compute cosine similarity against live genre centroids in tag-space. 3. **AI Explanations:** The top 5 mathematically closest genres are evaluated by **Gemini 2.0 Flash**, which writes an explanation of *why* it fits your specific taste profile. 4. **Visual Proof:** A stunning D3.js connection map draws the lines between what you know and what you're about to discover. <br /> ## 🚀 Features - **Data-Driven Discovery:** Recommendations are backed by real listener data (Last.fm), not cultural assumptions or AI inference. - **AI-Powered Insights:** Uses **Google Gemini 2.0 Flash** purely for the explanation layer. - **Visual Knowledge Graph:** Animated D3.js network visualization showing exactly how your inputs map to the new genres. -