---
title: Audio Restoration - SAM Audio
emoji: 🎵
colorFrom: indigo
colorTo: purple
sdk: gradio
sdk_version: 5.9.1
app_file: app.py
pinned: false
license: mit
python_version: "3.11"
---

# 🎵 Audio Restoration - Intent-Based AI

AI-powered audio restoration using **SAM-Audio** (Meta AI) for intent-based source separation.

## Features

- **Intent-Based Separation**: Describe what you want to isolate (speech, music, etc.)
- **Multiple Modes**: Basic separation, enhanced with span detection, or full restoration
- **Presets**: Optimized for Speech, Music, Field Recording, Historical Archives
- **Fallback**: noisereduce for compatibility when SAM-Audio unavailable

## Usage

1. Upload your audio file
2. Select restoration mode
3. Choose a preset for your audio type
4. (Optional) Describe what sound you want to isolate
5. Click "Start Restoration"
6. Download the restored audio

---

Built for SEAPAVAA Archivists 🏛️ | Powered by SAM-Audio (Meta AI)