SamurAIGPT/Text-To-Video-AI

What it solves

This project automates the creation of engaging short-form videos (like YouTube Shorts, Instagram Reels, and TikToks) from simple text prompts. It eliminates the manual effort of writing scripts, recording voiceovers, finding background footage, and syncing captions.

How it works

The system orchestrates several AI services to build a video end-to-end:

1. Scripting: Uses LLMs (OpenAI, Groq, or Google Gemini) to generate a script based on a user-provided topic.
2. Voiceover: Converts the script to speech using EdgeTTS (free) or ElevenLabs.
3. Visuals: Fetches relevant B-roll footage from Pexels or generates AI video segments via the MuAPI (supporting models like Google Veo, Sora, and Kling).
4. Captions: Uses Whisper or Deepgram to generate timestamps for the speech, creating synchronized on-screen text.
5. Assembly: Combines these elements into a final video file in either portrait or landscape orientation.

Who it’s for

Content creators, social media managers, and educators who want to produce high-quality, AI-driven video content quickly without needing professional editing software or manual asset sourcing.

Highlights

• Flexible AI Stack: Support for multiple LLM providers and high-end video generation models via MuAPI.
• B-Roll Integration: Automatic fetching of stock footage from Pexels.
• Customizable Styling: Full control over caption fonts, colors, and positions.
• Multiple Formats: Native support for both 9:16 (portrait) and 16:9 (landscape) orientations.
• Accessible Entry: Available via local installation, Google Colab, or a production-ready Premium API.