AI-Video-Maker

This project aims at creating a tool that helps to convert web contents into summarized videos to simplify work. Given a URL it is able to download the content of the page, generate a brief summary of the text as well as add narration using text-to-speech functionality. It also creates visual images known as B-roll according to the content of the page as well as provide written text for subtitles or texts overlay.

Features

• Web Page Parsing: Automatically extracts and summarizes web page content.
• Text-to-Speech: Converts the summarized text into a voiceover using external services or libraries.
• B-roll Image Generation: Generates relevant images based on the page content.
• Subtitles: Adds subtitles or text overlays based on the summarized content.
• Video Creation: Uses FFmpeg to combine voiceover, images, and subtitles into a complete video file.

Requirements

• Node.js
• FFmpeg (with ffmpeg-static and fluent-ffmpeg) - It is an open source multimedia tool that is used for capturing, decoding/encoding or streaming videos and audios, and other related tools with support for numerous formats.
• External text-to-speech and image generation services

Dependencies

• @gptscript-ai/gptscript: For generating summaries or interacting with AI models.
• ffmpeg-static and fluent-ffmpeg: To manage and process media content.
• express and cors: For handling HTTP requests and cross-origin resource sharing.
• dotenv: For managing environment variables.
• uniqid: To generate unique identifiers for files or requests.
• path and fs: For file system and path handling.

Installation

1. Clone the repository:

   git clone https://github.com/richayadav03/AI-Video-Maker.git
   cd AI-Video-Maker

2. Install the dependencies:

   npm install

3. Ensure FFmpeg is installed and available on your system.

Usage

1. Run the script by providing the URL of the webpage to convert:

   node generateVideo.js <webpage_url>

License

MIT License