Summary

myAiVideos provides an end-to-end automated pipeline for producing Chinese-language short videos, from topic discovery to distribution. It targets users seeking efficient, automated content creation, delivering publish-ready videos with a single command.

How It Works

The system employs a 7-layer architecture orchestrated by FastAPI, with a Java Spring Boot gateway managing platform governance. Key innovations include multi-model failover and circuit breakers (Resilience4j) for AI services, AI quality gating (prompt anchoring, CLIP consistency, AV sync rescue), and full-stack observability via Langfuse. This approach ensures robustness, consistency, and efficient resource utilization.

Quick Start & Requirements

• Prerequisites: Docker Desktop ≥ 24, Node.js ≥ 20 (frontend dev), Python ≥ 3.11 (local/tests).
• Installation: Clone the repo, configure .env with necessary AI provider API keys, and run docker compose up -d postgres redis orchestrator worker. Database migrations are required: docker compose exec -T orchestrator alembic upgrade head. The Java gateway and Vue 3 frontend can be started separately.
• Dependencies: API keys for DeepSeek, Zhipu GLM, Kling AI, Volcengine/MiniMax TTS; PostgreSQL, Redis, MinIO.
• Documentation: architecture.md, docs/operations-manual.md, docs/启动-停止脚本.md.

Highlighted Details

• Decoupled 7-layer pipeline architecture (Topic, Creative, Visual, Audio, Post-production, Distribution, Optimization).
• Robust AI model integration with multi-model failover and circuit breaker patterns.
• Advanced AI quality governance: prompt anchoring, CLIP text-image consistency, AV sync auto-rescue.
• Comprehensive observability using Langfuse and cross-language trace context.
• Production-ready features: SSE progress streaming, single-segment regeneration, rolling log archives.

Maintenance & Community

No specific community channels (Discord/Slack) or detailed contributor information are provided in the README. The project integrates several well-known open-source and commercial AI models.

Licensing & Compatibility

• License: MIT.
• Compatibility: The MIT license permits commercial use and integration into closed-source projects.

Limitations & Caveats

The pipeline is primarily designed for Chinese-language content. Setup requires obtaining and configuring API keys for multiple third-party AI services, which may incur costs and add complexity.