Update README.md

README.md

@@ -1,23 +1,72 @@
+---
+title: AI Song Identifier (Dejavu)
+emoji: 🎧
+colorFrom: purple
+colorTo: blue
+sdk: gradio
+sdk_version: "4.36.1"
+python_version: "3.10"
+app_file: app.py
+pinned: false
+---
+
 # 🎧 AI Song Identifier (Dejavu)
 
-This Hugging Face Space allows you to identify songs using audio fingerprinting.
+Identify songs using audio fingerprinting — similar to Shazam.
+
+---
+
+## 🚀 How it Works
+
+1. Upload or record an audio clip  
+2. The system generates an audio fingerprint  
+3. Matches it against a fingerprint database  
+4. Returns the best match with confidence  
+
+---
 
-## 🚀 How it works
-- Audio is converted into fingerprints
-- Matched against a fingerprint database
-- Returns best match
+## 🧠 Tech Stack
 
-## ⚠️ Notes
-- Only works for songs in the database
-- Database is updated via GitHub automation
+- Gradio (UI)
+- Dejavu (Audio Fingerprinting)
+- SQLite (Fingerprint Database)
 
-## 🛠 Tech Stack
-- Gradio
-- Dejavu (audio fingerprinting)
-- SQLite database
+---
 
 ## 📂 Database
-This Space uses `fingerprints.db` (no audio stored)
+
+- This app uses `fingerprints.db`
+- Only fingerprints are stored (no audio files)
+- Database is updated automatically via GitHub Actions
+
+---
+
+## ⚠️ Limitations
+
+- Works only for songs present in the database  
+- Accuracy depends on audio quality  
+- Background noise may affect results  
+
+---
 
 ## 🔄 Updates
-Fingerprint database is automatically updated via GitHub Actions.
+
+- Songs are automatically:
+  - scraped (externally)
+  - fingerprinted
+  - pushed to this Space  
+
+---
+
+## 🛠 Future Improvements
+
+- Neural audio fingerprinting (FAISS)
+- Faster search & scaling
+- Better noise robustness
+
+---
+
+## 💡 Note
+
+This project is for educational and experimental purposes.
+No copyrighted audio is stored in this repository.