Skip to content

oprogramadorreal/audara-studio

v0.14.1MIT

Make videos from code: a deterministic WebGL/GLSL, three.js and Canvas2D engine with a live browser preview you direct by time and an offline render to MP4, plus a soundtrack skill that turns a song or a script into audio and timing data (and a narration into dubbed tracks in other languages), a video-ideas skill that helps choose what to make and writes the starting prompt or script, a devlog skill that builds something worth watching being built and keeps the material for the video about how it was made, a video-repo skill that puts a project in git for someone else to clone and render, a youtube-package skill that makes a finished video's thumbnails, title, description, tags and subtitles, and the same in more languages, and a nutlight-style skill, a house style for explainers (squirrel characters in luminous geometric worlds) that loads when it is named.

soundtrack

Turns a song or a script into a video's soundtrack plus the timing data the picture syncs to: beats, downbeats, sections and word timings as JSON, narration and music generated with ElevenLabs (asking before spending credits), sound effects, the video's mix of voice, music and effects, and its narration dubbed into other languages. Use it whenever a video or animation, made with code, filmed or AI-generated, needs its music analyzed, a voiceover made, lyrics or words timed, background music or effects created, its audio mixed or its narration dubbed, even if the user only says "make it sync" or "add a voiceover". Not for podcasts, making a voice clone, noise removal, mastering a song for streaming, or app code that calls a speech API.

Read SKILL.md at the source

Pinned to revision 7fe01e0d8125, so it is the text this page describes rather than whatever the author pushed since.

Files

Every link opens the file at its source, pinned to the revision this page describes.