Podcast2Newsletter

Podcast2Newsletter is a Python-based tool that automates the process of converting podcast episodes into newsletters. It downloads podcast audio, transcribes it, summarizes key points, and formats them into a Markdown newsletter.

---

Features

• Podcast Download: Fetches podcast episodes from an RSS feed.
• Audio Processing: Splits audio files into manageable chunks using FFmpeg.
• Transcription: Converts audio to text using the Whisper ASR model.
• Summary and Newsletter Generation: Summarizes transcriptions and formats them into a newsletter in Markdown.
• Handlebars Template: Provides an easy way to customize newsletter formatting.

---

Requirements

• Python 3.7+
• FFmpeg
• OpenAI Whisper
• Google Generative AI API (Gemini)
• Required Python Libraries:
  • feedparser
  • whisper
  • dotenv
  • pybars
  • tqdm
  • google-generativeai

---

Installation

1. Clone the repository:

   git clone https://github.com/your-username/podcast2newsletter.git
   cd podcast2newsletter

2. Install the dependencies:

   pip install -r requirements.txt

3. Ensure FFmpeg is installed and its path is set in the .env file.

---

Configuration

Create a .env file in the root directory and add the following environment variables:

PODCAST_URL=<Your Podcast RSS Feed URL>
FFMPEG_PATH=<Path to FFmpeg executable>
GEMINI_API_KEY=<Your Google Gemini API Key>

---

Usage

1. Run the script:

   python podcast2newsletter.py

2. The script will:

   • Download the latest podcast episode from the RSS feed.
   • Split the audio into chunks.
   • Transcribe each chunk.
   • Generate a Markdown newsletter summarizing the episode.

3. Output files:

   • Transcriptions: Stored in chunks/transcriptions/.
   • VTT Files: Stored in chunks/vtt/.
   • Final Newsletter: newsletter.md

---

How It Works

1. Podcast Download:

   • Parses the RSS feed and downloads the latest episode audio.

2. Audio Chunking:

   • Splits the audio into smaller chunks for easier processing.

3. Transcription:

   • Transcribes each chunk using Whisper ASR.

4. VTT Merging:

   • Merges transcriptions into a single VTT file.

5. Summary Generation:

   • Generates a summarized Markdown newsletter using Google Gemini API and Handlebars templates.

6. Markdown Output:

   • The final newsletter is saved as newsletter.md with clickable timestamps.

---

Example Output

# Episode Title

Episode Summary

## Section Header

Section content.

[00:05:30](<Podcast Episode URL>#t=330)

## Another Section Header

More content.

[00:10:15](<Podcast Episode URL>#t=615)

---

License

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

---

Contributions

Contributions are welcome! Feel free to fork the repository and submit a pull request.

---

Star History