@allyhub/analysis
v0.1.0-alpha.8
Published
AllyHub's media analysis on this computer: speech recognition with word times (Parakeet), shots, onsets and tempo, on-screen text (OCR), face tracks (YuNet) and person mattes (U²-Net), contact sheets, reference ads and their comparison with a render; mode
Readme
@allyhub/analysis
What AllyHub learns from media files, on this computer: speech with word times, cuts, beats and tempo, on-screen text, face tracks and person mattes, and the offset between two recordings. Models and runtimes download the first time they are needed.
This package is part of AllyHub: product demo videos from a few sentences, made by your coding agent. You do not install it yourself. The allyhub command brings it, and a project folder needs no package.json.
To use AllyHub, add the skill to your agent:
npx skills add allyhub-ai/skills --skill allyhub -g -a claude-code codexLicense
Proprietary: you may install and use it as set out in LICENSE. Code adapted from other projects is listed in NOTICE. "AllyHub" is a trademark of AllyHub, Inc.
