# Audio2Spec

**Repository Path**: mantis3d/Audio2Spec

## Basic Information

- **Project Name**: Audio2Spec
- **Description**: A Python-based desktop tool to batch convert audio files (WAV, MP3, FLAC, etc.) into Mel-spectrogram images with a 1:1 filename mapping. 
- **Primary Language**: Python
- **License**: Not specified
- **Default Branch**: main
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2026-06-02
- **Last Updated**: 2026-06-02

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# Audio to Spectrogram Batch Converter

🚀 A Python-based desktop tool to batch convert audio files (WAV, MP3, FLAC, etc.) into Mel-spectrogram images with a 1:1 filename mapping. It's perfect for dataset preprocessing in deep learning audio classification or acoustic analysis.

---

## ✨ Features

* **📦 Batch Processing**: One-click folder scanning that supports `.wav`, `.mp3`, `.flac`, `.ogg`, and `.m4a` formats.
* **🎨 Real-time Preview**: Instantly view the spectrogram effect on the right when switching between different color themes (like viridis, magma, gray, etc.) for easy comparison.
* **⚙️ Highly Customizable**:
  * Supports custom image resolution (DPI) to meet needs ranging from quick previews to academic paper figures.
  * Auto-crops edges: generated images have **no coordinate axes and no white borders**, tailored purely for feature extraction.
* **⚡ Thread Optimization**: Multi-threading ensures the UI never freezes when converting large batches of files, accompanied by real-time progress bar feedback.
* **🖥️ Minimalist GUI**: Built on Tkinter, eliminating the need for complex command-line operations.

---

## 🛠️ Prerequisites

Before running or developing, please make sure you have Python 3.x installed on your computer, along with the following dependency libraries:

```bash
pip install librosa matplotlib numpy Pillow
```

---

## 🚀 Quick Start  

### 1. Prepare Preview Assets (Must-read for first-time use)
To enable the UI to display color theme previews, you need to generate a set of assets first:
* Place a short audio file named `sample.wav` in the project root directory.
* Run the asset generation script, or ensure there is a `previews` folder in the root directory containing pictures for different themes like `preview_viridis.png`.

### 2. Start the Program
Run the main script in your terminal:
```bash
python audioToSpectrogram.py
```

### 3. Set Paths
* Click the **Browse** button to the right of "Audio Directory" to select the folder containing your audio files.
* Click the **Browse** button to the right of "Image Save Directory" to select the folder where you want to save the spectrograms.

### 4. Adjust Parameters & Preview
* Modify the image resolution (DPI) in the "Parameters" section.
* Toggle the "Color Theme" dropdown menu to **preview** the visual effects of different themes in real time on the right.

### 5. Start Conversion
Click the **Start Batch Conversion** button at the bottom. The progress bar will show the real-time progress. A prompt box will pop up upon completion.

---

## 📄 License

This project is licensed under the [MIT License](LICENSE). You are free to use, modify, and distribute this code.