BELIEVEYOUME, Synced Lyrics Demo

Shiner / The Life and Times. Words highlight from the forced-alignment timing, the pipeline output. Ten songs, word-level karaoke sync, no audio needed for the demo.

This is the automated output: canonical lyrics + per-word timestamps (enhanced LRC + a JSON sidecar). Held vowels stay one word here (the karaoke standard); the per-note melisma split is a separate tab layer. Anti-cascade: each line is pinned to its own timestamp, so one bad line cannot shift the rest.

Download the files (real links)

Per song: .lrc = enhanced karaoke lyrics with per-word timestamps, .json = structured sidecar, syllables = the plain paste text.

Song	LRC	JSON	Syllables
Asleep In The Trunk	.lrc	.json	-
So Far So	.lrc	.json	syllables
The Alligator	.lrc	.json	syllables
Broken Satellites	.lrc	.json	syllables
Lazarus	.lrc	.json	syllables
Not Too Much	.lrc	.json	syllables
Endless Summer	.lrc	.json	syllables
The Mutiny	.lrc	.json	syllables
My Mirror Hates Me	.lrc	.json	syllables
Jackie	.lrc	.json	syllables

Skill code and docs: run.py install.sh SKILL.md Asleep tab placement + confidence flags