YouTube Transcript Extractor (ytx)

A command-line tool to fetch and save YouTube video transcripts in plain text or SRT format, using yt-dlp and supporting cookies for private or age-restricted videos.

Features

• Extracts transcripts from YouTube videos (supports both manual and auto-generated captions)
• Output formats: plain text (.txt) or SRT (.srt)
• Handles cookies from browser or file for restricted videos
• Saves output to the output/ directory

Installation

1. Clone the repository:

   git clone https://github.com/juliustnt/transcript_scraper.git
   cd transcript_scraper

2. Install dependencies:

   pip install -e .

   This will install yt-dlp and requests as required dependencies.

Usage

Run the CLI tool with:

ytx-transcript <YouTube URL or ID> [plain|srt] [--cookies <cookies.txt>] [--browser <browser>] [--no-browser-cookies]

Arguments

• <YouTube URL or ID>: The video to fetch the transcript from
• [plain|srt]: Output format (default: plain)
• --input <PATH>: Path to a text file containing multiple YouTube URLs/IDs (one per line). Mutually exclusive with <YouTube URL or ID>
• --cookies <cookies.txt>: Path to a cookies.txt file (Netscape format)
• --browser <browser>: Try to use cookies from a browser (edge, chrome, firefox, brave, vivaldi, chromium)
• --no-browser-cookies: Never attempt to use browser cookies

Example

Fetch a transcript in plain text:

ytx-transcript https://www.youtube.com/watch?v=dQw4w9WgXcQ

Fetch a transcript in SRT format and save using Chrome cookies:

ytx-transcript https://www.youtube.com/watch?v=dQw4w9WgXcQ srt --browser chrome

Output

• Transcripts are saved in the output/ directory, named after the video ID and format (e.g., dQw4w9WgXcQ.txt or dQw4w9WgXcQ.srt).

Project Structure

transcript_scraper/
├── pyproject.toml
├── README.md
├── output/
└── src/
    └── ytx/
        ├── __init__.py
        ├── captions.py
        ├── cli.py
        ├── io_utils.py
        └── parse.py