Mxx

🎬 Mxx is a user-friendly video subtitle generation tool that uses OpenAI's Whisper model to detect audio content and automatically generate subtitles.

In addition to subtitle generation, Mxx also provides an AI-powered summarization feature, allowing users to quickly generate concise summaries of the video content based on the transcribed text.

✨ Features

• 🗣️ Automatic Speech Recognition (ASR)
  Transcribes spoken audio from videos into text using high-accuracy Whisper models.

• 🤖 AI-Powered Summarization
  Generates intelligent summaries based on the transcribed content, helping users quickly grasp the main ideas of the video.

• 🎞️ Multi-Format Video Support
  Supports common video formats including MP4, MOV, MKV, and more.

• 📝 Flexible Subtitle Formats
  Export subtitles as plain text or in the ASS subtitle format.

🚀 How to Use

Note for Linux: Make sure ffmpeg is installed and available in your system path.
Note for Windows: Place ffmpeg.exe in the same directory as Mxx.exe.

🌐 Web Interface

You can start the web interface in two ways:

1. Double-click (Windows only):
   Simply double-click the Mxx.exe file to launch the built-in web server.

2. Via command-line:

   Mxx --web

This will start both the front-end and backend API on the same server.

By default:

• Frontend: http://localhost:8080
• Backend API: http://localhost:8080/api

🌐 API Server and Front-End (Separate Deployment)

1. Start the backend API:

Mxx --api

2. Create a .env file inside the web/ directory with the following content:

   VITE_REACT_APP_API_URL=http://localhost:8080
   

3. Start the frontend:

   npm install
   npm run dev

This will start a local development server.

• Frontend: http://localhost:5173
• Backend API: http://localhost:8080

🧾 Command-Line Interface

To view available CLI options:

mxx --help

Use CLI commands to process video files directly from the terminal.

🔧 Build Guide

💡 Note: It is recommended to compile the Whisper library yourself to enable more features such as GPU acceleration.
Otherwise, the default prebuilt version in the whisper/lib folder will be used (GPU is not enabled by default). 💡 Note: See how to enable GPU for whisper : windows

🪟 Notes for Windows

• The prebuilt Whisper library for Windows is located at:
  whisper/lib/win

• If you encounter the error 0xc0000139 during runtime,
  make sure to place the correct libstdc++-6.dll into the whisper/lib/win folder.

• You need to install MinGW, and ensure make and the required toolchains are available in your system PATH.

---

🛠️ Build Command

To build the project from source:

make build

• The backend executable will be generated as: Mxx
• The frontend static files will be output to the dist/ folder