Timeline Studio β Local-first Browser AI Video Editing
Collection
Open-source browser video editing with local ONNX captions, voiceovers, source separation, vision tools, and talking avatars. β’ 2 items β’ Updated
How to use haixin/timeline-studio-onnx-models with LivePortrait:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
Browser-compatible ONNX model assets used by Timeline Studio, an MIT-licensed, local-first AI video editor.
The repository contains revision-pinned model assets for Timeline Studio's browser inference workflows. Large downloads are cached by the editor's service worker so repeat use does not fetch the same model files again.
liveportrait-webgpu/ β LivePortrait browser animation models.joyvasa/ β JoyVASA audio-driven motion models.migan-webgpu/ β MI-GAN 256 ONNX adaptation for local object and watermark
removal. See the directory README and bundled upstream MIT license.nanovsr-644k/ β NanoVSR 644K FP16 ONNX exports for 4Γ image and temporal
video restoration. See the directory README and bundled upstream MIT license.Each adapted model directory documents its upstream project, browser
conversion, intended use, limitations, and artifact checksum. Timeline Studio
loads production assets from immutable commit revisions rather than mutable
main URLs.