Use Python to convert videos into high-quality little red book notes with one click

By /

Automatically optimize content and mapping
This is a way to use Python and AI to extract and optimize the footsteps of the Little Red Book. The code is open source

Video_note_generator is an open source project developed by GitHub user whotto. It aims to convert video content into high-quality little red book notes with one click, and automatically optimize content and accompanying pictures.

Application scenarios:

  • Content creators: Quickly convert video or live content into articles to improve creative efficiency.
  • Knowledge manager: Automatically organize video notes and learning points to facilitate knowledge accumulation.
  • Social media operations: Generate high-quality Little Red Book notes in batches to enhance the influence of social media.

    Characteristics of Little Red Book Notes:

Title Creation:

  • Use the diode header method to capture the user’s pain points.
  • Integrate high-conversion words and emotional words.
  • Add 2-4 related emoticons.
  • Control it within 20 words.

    Content optimization:

  • The text is controlled at 600-800 words.
  • Use emoji to guide each paragraph.
  • Set up 2-3 interactive guides.
  • Add personal experience and empathy descriptions.
  • Automatically obtain relevant drawings.

    Label system:

  • Core keywords.
  • Associated long-tail words.
  • High conversion label.
  • Hot topic tags.

    Creation process:

  • Enter the video URL.
  • Download the video and extract the audio.
  • Use Whisper to transcribe audio content.
  • Organize long text content through AI.
  • Optimize it to the small red book style, including generating titles, tags and obtaining accompanying pictures.
  • Generate final notes.

    Usage:

Three usage methods are supported:

Processing a single video:

python video_note_generator.py https://example.com/video

Batch processing URL files:

Place each video link in the urls.txt file, one link per line, and run:

python video_note_generator.py urls.txt

Processing Markdown files:

Support Markdown files with video links:

python video_note_generator.py notes.md

Use tools:

  • FFmpeg: Used for audio-video conversion.
  • Whisper: Used for speech to text.
  • OpenRouter: Used for AI content optimization.
  • Unsplash: Used to obtain high-quality pictures.

    Quick Start:

  • Installation dependencies:
  • Install FFmpeg.
  • Install Python dependencies:
pip install -r requirements.txt

Configure environment variables:

cp .env.example .env

Configure API keys:

Edit the.env file and fill in the necessary API keys:

OPENROUTER_API_KEY=your-api-key-here
UNSPLASH_ACCESS_KEY=your-unsplash-access-key-here
UNSPLASH_SECRET_KEY=your-unsplash-secret-key-here

Start using:

Create a urls.txt file with one video link per line.

Operating environment inspection:

python check_environment.py

Run the generator:

python video_note_generator.py test.md

Output file:

Each video generates three files:

  1. Original notes (YMMDD_HHMMSS.md):

    • Complete video transcribed text, retaining all details.

  2. Organize notes (YMMDD_HHMMSS_organized.md):

    • The structured content after AI optimization will be highlighted with highlights and paragraphs optimized.

  3. Little Red Book version (YMMDD_HMMSS_xiaohongshu.md):

    • Optimized hit title.
    • The essence of 600-800 words.
    • 2-3 Zhang related drawings.
    • Optimized labeling system.
    • Interactive guide design.

      Configuration instructions:

The following parameters can be adjusted in the.env file:

  • MAX_TOKENS: The maximum length of the content of the small red book generated.
  • CONTENT_CHUNK_SIZE: Long text chunk size (number of characters).
  • TEMPERATURE: AI Creativity Level (0.0 – 1.0).

License:

The project uses an MIT license.

Github：https://github.com/whotto/Video_note_generator

Oil tubing:

Scroll to Top