What problem does it solve? Audio tracks placed in a HyperFrames composition often fight each other — a music bed drowns out a voiceover, a voice sounds boomy or harsh, or levels drift across a take. This Skill mixes already-placed audio by writing effect chains, volume automation, and voiceover carve settings directly onto the composition's HTML elements, with preview and render running the same Web Audio graph. ## Core Features & Use Cases - Voiceover carve: Analyzes voice tracks and writes peaking-filter dips plus a level-match envelope onto the music bed via scripts/carve.mjs, so the bed keeps its character while speech stays intelligible. - Effect chains and automation: Fourteen shipped effects (EQ, compressor, limiter, gate, saturation, delay, reverb, chorus, phaser, bitcrush) written as data-fx-chain JSON, with breakpoint automation lanes on volume and automatable parameters. - Diagnosis and presets: Reference guides for measuring audio defects with ffmpeg when you cannot listen, plus named presets and jobs (Tame Boominess, Reduce Mud, Add Clarity) mapped to symptoms. - Use Case: Given a composition with a narration track and a music bed, run the carve script to automatically detect both tracks, analyze the voice's bands, and write a dynamic carve that ducks the bed only where speech occurs. ## Quick Start Run the carve script against my HyperFrames composition index.html to automatically duck the music bed under the narration.
