Viddie - AI-Powered Video Creation

Short Description

Viddie is an AI-powered platform for automated video creation, utilizing advanced machine learning models. It streamlines video production by combining GPT-4 for script generation, Conquis TTS for speech synthesis, DALL-E for image generation, and SadTalker for avatar animation. This allows users to generate high-quality videos with minimal manual effort.

Frontend Repository

Frontend Repository

Sample Videos

• Demo Video 1
• Demo Video 2

---

How to Run the Project

There are two ways to run the project: manually or using Docker.

Manual Installation

Prerequisites

• Recommended Python version: 3.9 - 3.11
• Install the following dependencies:
  1. FFmpeg (Required for video rendering) - Installation Guide
  2. ImageMagick (Required for subtitles) - Download
  3. eSpeak-NG (Only required for some voice models) - Download MSI
     • If you don't want to install extra voice packages, use only:

       tts_models/en/ljspeech/tacotron2-DDC
       

Installation Steps

1. Navigate to the viddie folder and install dependencies:

   pip install -r requirements/requirements.txt

2. Create a checkpoints folder and download required files from: Google Drive - Checkpoints

3. Inside the viddie folder, create a .env file and add the following API keys:

   • OPEN_API_KEY
   • SEARCH_ENGINE_ID
   • API_KEY

   To find your Google search engine ID and API key, refer to this YouTube Guide.

4. Run the following commands to set up the database and start the server:

   py manage.py makemigrations
   py manage.py migrate
   py manage.py loaddata fixtures/fixtures.json
   py manage.py setup_media
   py manage.py runserver

5. (Optional) To enable ElevenLabs voices, add your XI_API_KEY in the .env file and run:

   py manage.py setup_elevenlabs

---

Docker Installation

1. Create the .env file and add your OPEN_API_KEY, SEARCH_ENGINE_ID, API_KEY as per .env_example.
2. Download and place the required checkpoint files in the checkpoints folder from: Google Drive - Checkpoints
3. Navigate to the src folder and run:

   docker-compose up --build

---

Admin Panel

• URL: http://localhost:8000/admin
• Login Credentials:
  • Username: admin
  • Password: pass

API Documentation

You can find all API endpoints in Swagger: http://localhost:8000/swagger/

---

Roadmap / To-Do List

• Convert all moviepy functions to FFmpeg for better performance.
• Add Celery support for asynchronous and scheduled tasks.
• Implement unit tests for models, functions, and views.

---

Contact & Support

For any inquiries or support, feel free to reach out:

• University Email: kodamia@cs.ihu.gr
• Personal Email: kostas2372@gmail.com

---

Recent Updates

✅ Fixed Conquis TTS Docker issue
✅ Added support for Gemini and Claude AI models
✅ Integrated ElevenLabs API voices
✅ Enabled compilation video creation from Twitch (by game or streamer)
✅ Added OpenAI voices
✅ Integrated MidJourney and Stable Diffusion as image providers (Change providers in ``)
✅ Dockerized the application for easier deployment

---