This project provides both a web-based (Streamlit) and command-line (CLI) tools to convert YouTube video transcripts into easily readable Markdown, Plain Text, or JSON formats.

• 🔗 Supports YouTube video URLs, IDs, and Playlists (batch processing CLI).
• 🌍 Multiple language support for transcripts.
• 📊 Export options: Markdown, Plain Text, and JSON.
• 📁 Batch processing from a file of URLs/IDs (batch_processing_yt.py).
• 🎨 Dark mode toggle for better UX (Streamlit app).
• 📑 Automatic paragraph formatting for readability.
• 👁️ Transcript preview (Streamlit app).
• 📂 User-specified output directory.
• ✅ Sanitized, Linux-friendly, kebab-case filename generation based on video title.

1️⃣ Create and activate a virtual environment:

# Conda (recommended)
conda create -n youtube-md python=3.10
conda activate youtube-md

# OR using venv
python -m venv .venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate

2️⃣ Install UV (recommended for faster installation):

pip install uv

3️⃣ Clone this repository:

git clone https://github.com/ZeroXClem/YouTube-to-Markdown.git
cd YouTube-to-Markdown

4️⃣ Install dependencies:

# Using UV
uv pip install -r requirements.txt

# Or pip
pip install -r requirements.txt

Note: If any issues arise, install these individually: youtube-transcript-api, yt-dlp, requests.

Run the Streamlit app:

streamlit run youtube-transcript-downloader2.py

Then in your browser:

• Enter a YouTube video URL or ID.
• Select your language and format.
• Click Download Transcript.

1️⃣ Run:

python youtube_cli.py

2️⃣ Follow prompts to:

• Enter video URL/ID.
• Choose transcript language.
• Pick export format (md, txt, json).
• Specify output directory.

✅ Features:

• Step-by-step guided experience.
• Auto-formatted paragraphs.
• Clean Linux-friendly, kebab-case, ASCII-only filename generation.

For multiple videos or playlists:

python batch_processing_yt.py <input> [-l language] [-f format] [-o output_directory]

Arguments:

• input (required): Single video URL/ID, playlist URL, or text file with multiple URLs/IDs.
• -l, --language: Language code (default: en).
• -f, --format: Output (txt, md, json, default: txt).
• -o, --output: Output directory (default: current).

Examples:

✅ Single Video:

python batch_processing_yt.py https://www.youtube.com/watch?v=dQw4w9WgXcQ -l en -f md -o my_transcripts

✅ Playlist:

python batch_processing_yt.py "https://www.youtube.com/playlist?list=PLQVv..." -l en -f json -o playlist_transcripts

✅ From a File:

video_list.txt contents:

https://www.youtube.com/watch?v=dQw4w9WgXcQ
another_video_id
https://www.youtube.com/playlist?list=PLQVv...

Run:

python batch_processing_yt.py video_list.txt -l fr -f md -o my_transcripts

✅ Help:

python batch_processing_yt.py --help

├── .git/
├── requirements.txt
├── transcript_helper.py
├── versions/
├── batch_processing_yt.py
├── youtube_cli.py
└── youtube-transcript-downloader2.py

• requirements.txt: Python dependencies.
• transcript_helper.py: Shared helper functions.
• batch_processing_yt.py: CLI for batch processing.
• youtube_cli.py: Interactive CLI with Linux-friendly filename sanitation.
• youtube-transcript-downloader2.py: Streamlit web app.

Contributions, issues, and feature requests welcome!
Check the issues page.

MIT License. See LICENSE.

• Streamlit for the app framework.
• youtube-transcript-api.
• yt-dlp for reliable video and playlist info.

Made with ❤️ by [ZeroXClem]