Summary

AI Media Assistant is a local, web-based tool for Chinese content creators, automating short video generation. It integrates text editing, subtitle styling, automatic image and BGM selection, text-to-speech (TTS) narration, and video export within a browser, keeping all project data and generated outputs stored locally for privacy and control.

How It Works

The project utilizes a React/TypeScript/Vite frontend and a FastAPI backend. Core media processing, including subtitle layout, template application, TTS integration, and video synthesis, is handled by a dedicated media_core package. Background worker processes manage computationally intensive TTS and video rendering tasks, ensuring API responsiveness. The system supports multiple TTS engines and features an intelligent image-matching system for emotional video templates, drawing from local resource libraries.

Quick Start & Requirements

• Installation: Download the Release zip. Windows users run scripts/setup_windows.ps1 then scripts/start_windows.ps1. macOS users run chmod +x scripts/setup_macos.sh scripts/start_macos.sh followed by ./scripts/setup_macos.sh.
• Access: http://127.0.0.1:8123
• Prerequisites:
  • OS: Windows 10/11 or macOS.
  • Python: 3.11+ (ensure separate installation on macOS if system Python is older).
  • Node.js: 20+.
  • FFmpeg: Recommended to be added to system PATH.
  • Model Weights: Required for TTS (OmniVoice, Qwen3-TTS) and must be downloaded and configured separately by the user.
• Documentation: docs/USER_GUIDE.md

Highlighted Details

• Local-First Operation: All project data and generated videos are stored locally, enhancing user privacy and control.
• Automated Content Matching: Features automatic background image selection for emotional video templates based on script content.
• Multi-Engine TTS: Supports various text-to-speech engines, including OmniVoice and Qwen3-TTS, with configurable voice and speed.
• Customizable Subtitle Templates: Offers diverse subtitle styles like centered large text, scrolling queues, and thematic templates.

Maintenance & Community

The provided README does not detail specific community channels (e.g., Discord, Slack), notable contributors, or sponsorship information. It directs users to docs/USER_GUIDE.md for comprehensive usage instructions.

Licensing & Compatibility

The project's license type is not specified in the provided README. This omission requires further investigation for commercial use or integration into closed-source projects.

Limitations & Caveats

Model weights for TTS engines are not included and require manual acquisition and configuration. Model runtime environments are platform-specific; Windows and macOS model packages are incompatible. The tool is primarily tailored for Chinese language content creators. The project's license is not disclosed, posing a potential adoption blocker.