MartinDelophy/ai-video-editor

Open-source, local-first video editor where creators and AI agents edit the same real timeline.

What it solves

Timeline Studio provides a local-first, browser-based video editing suite that integrates AI capabilities directly into a multi-track timeline. It eliminates the need to upload project media to a backend for AI processing, ensuring privacy and reducing latency while providing professional-grade editing tools.

How it works

The editor runs in the browser using WebGPU and WASM for local inference. It leverages ONNX models for tasks like music generation and speech-to-text, and uses WebCodecs for MP4/WebM composition. Models are lazy-loaded and cached via service workers, with mirrors on Hugging Face and ModelScope to ensure availability. It also includes a headless command runner that allows AI agents to plan and execute edits on .timeline project files.

Who it’s for

It is designed for content creators and developers who want a CapCut-style editing experience with integrated AI tools (like talking avatars and automatic captions) that run locally in their browser.

Highlights

  • Local AI Suite: Includes multilingual voiceovers, Stable Audio 3 for music, Whisper for automatic captions, and MI-GAN for watermark removal.
  • Visual AI Tools: Features YOLOS for smart framing, MODNet for background removal, and LivePortrait for digital human animation.
  • Professional Timeline: Supports multi-track editing, keyframes, four-way color grading, and ripple editing.
  • Agent-Ready: Provides a dedicated "AI Video Editing Skill" allowing LLM agents to inspect and modify video projects programmatically.
  • Privacy-Focused: Local-first inference means media stays on the user's machine during the AI workflow.

Related

  • Project
    ErickWendel/localstudio

    A browser-native, local-first slide editor that uses Web AI and WebGPU to provide AI-powered slide generation, translation, and presentation tools without a backend.

  • Project
    open-ribbi/velocut

    An AI-native, local-first video editor that runs entirely in the browser, allowing both humans and AI agents to edit footage via a shared JSON command protocol.

  • Project
    Sportinger/MasterSelects

    A local-first, GPU-accelerated media editor for video, audio, 3D, and vector animations that integrates on-device AI for segmentation and timeline control.

  • Project
    0xsline/OpenChatCut

    An open-source, local-first AI video editor that combines a professional multitrack timeline with conversational agents for editable, agent-driven video production.

  • Project
    jianjieyiban/JJYB_AI_VideoAutoCut

    A local-first AI video creation workstation that automates the workflow from material analysis and scriptwriting to voiceover generation and audio-visual synchronization.