The corpus, in public
Every vote you cast in the Taste Test feeds the ground-truth corpus that calibrates the Lyric Scoring Standard. The numbers below are the live aggregate — open data, no login, no fingerprinting. The bigger the corpus, the stronger the calibration signal.
Total votes
0
pairwise judgments
Rubric agreement
—
0 agreed / 0 total
Last updated
—
aggregates date
Only 0 votes so far. The corpus needs more — every vote moves the calibration signal forward.
Agreement by score-gap band
When the two songs differ by a wider score gap, the rubric SHOULD agree with human listeners more often. That's the basic calibration property — the bigger the rubric's claimed difference, the more obvious it should be to a human ear.
| Gap (pts) | Votes | Agreed | Rate |
|---|---|---|---|
| 0 | 0 | 0 | — |
| 1-3 | 0 | 0 | — |
| 4-7 | 0 | 0 | — |
| 8-12 | 0 | 0 | — |
| 13+ | 0 | 0 | — |
Add to the corpus
The data above only exists because people contributed votes. Every additional vote tightens the agreement measurement and feeds the next quarterly calibration report. No login, no email — one click.
Vote in the Taste Test