FEDRA: Flexible, Easy, and Dynamic Recorder Automation

Overview

FEDRA is a lightweight tool designed to streamline the transcription of audio files using OpenAI Whisper. With integrated logging, a web server, and organized Markdown output, this program ensures a seamless workflow for researchers and professionals alike.

---

Features

• Automatic Transcription: Uses OpenAI Whisper for accurate transcriptions.
• Web Server: Easily access audio files via a built-in web server.
• Markdown Output: Save transcriptions in organized, timestamped Markdown files.
• Dynamic Folder Creation: Automatically organizes files into monthly subdirectories in your Obsidian Vault.

---

Device Options: CPU enabled by default.

• CPU: Use if no GPU is available or if GPU is not desired.
• CUDA: For Nvidia GPUs supporting CUDA (requires PyTorch with CUDA).
• MPS: For Apple Silicon devices (M1/M2 chips) using Metal Performance Shaders.
• OpenCL: If supported by hardware and libraries.
• DirectML: For Windows environments using DirectML (via ONNX Runtime or PyTorch DirectML backend).

Example:

import torch
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")

---

Installation

1. Clone the Repository

git clone https://github.com/your-repo/fedra.git
cd fedra

2. Setup a Virtual Environment

python3 -m venv whisper_env
source whisper_env/bin/activate

3. Install Dependencies

pip install -r requirements.txt

---

Usage

1. Prepare Your Input Folder:

Place audio files (.mp3, .wav, .m4a) into the designated input folder (default: Audio Hijack).

2. Run the Main Script:

python main.py

3. Access the Web Server:

After starting the program, visit the web server to access audio files:

http://localhost:8000

4. View Generated Markdown:

Markdown files are saved in the output folder and organized into monthly directories under your Obsidian Vault.

5. Stop Gracefully:

Use Ctrl+C to gracefully terminate the program.

---

Configuration

• Input Folder: Set the path to the folder containing your audio files (default: /Users/jhumphreys/Music/Audio Hijack).
• Output Folder: Set the path for the transcriptions (default: /Users/jhumphreys/Scripts/transcription/output).
• Obsidian Vault: Update the path to your Obsidian Vault for Markdown files (default: /Users/jhumphreys/Documents/Obsidian Vault/Meeting Notes).

You can modify these paths in the main.py file.

---

File Structure

fedra/
├── main.py           # Entry point of the program
├── transcription.py  # Transcription logic using Whisper
├── web_server.py     # Lightweight web server logic
├── file_utils.py     # Helper functions for file handling
├── logging_utils.py  # Logging setup and utilities
├── requirements.txt  # Python dependencies
├── output/           # Transcribed Markdown files
├── Audio Hijack/     # Default input folder for audio files
└── README.md         # Project documentation

---

Example Output

Markdown File

# Transcription: 2024-12-19 2051 Recording

## Overview
Transcription generated on 2024-12-19 20:51:00.

## Original Audio
[Original Audio File](http://localhost:8000/2024-12-19_2051_Recording.mp3)

## Content
[00:00:00 --> 00:01:00](http://localhost:8000/2024-12-19_2051_Recording.mp3#t=0) Sample transcription text...
[00:01:00 --> 00:02:00](http://localhost:8000/2024-12-19_2051_Recording.mp3#t=60) More transcription text...

Directory Structure

/Meeting Notes/
└── 12-2024/
    └── 2024-12-19_2051_Recording.md

---

License

This project is licensed under the MIT License - see the LICENSE file for details.

---

Contributing

We welcome contributions! Feel free to submit pull requests or open issues.

1. Fork the repository.
2. Create a feature branch: git checkout -b feature-name.
3. Commit your changes: git commit -m 'Add a feature'.
4. Push to the branch: git push origin feature-name.
5. Open a pull request.

---

Acknowledgments

• OpenAI Whisper for transcription capabilities.
• The Python community for tools like http.server, logging, and shutil.