Monorepo for Aesthetic.Computer aesthetic.computer
README.md

blackboard #

jeffrey sings Tim Ingold's blackboard/PowerPoint passage from "Thinking through Making" (Institute for Northern Culture lecture, 07:28–08:28 — clip and transcript in papers/lines-platter/). 72 BPM, E minor, 48 bars, 2:44. The hook — "that's why I like blackboards / and don't like PowerPoint" — lands at bars 20, 32, and 40.

Built bottom-up: the bed is synthesized in bin/render-bed.mjs (chalk-scrape percussion via the TrackDrum friction voice ported from pop/cult, patient E-minor pads, section dynamics intro→hooks ≈ −32→−18 dB RMS); the vocal is @jeffrey's ElevenLabs voice (jeffrey-pvc, stability 0.55) lifted to singing through the aesthetivox chain (spinging/ — continuous WORLD line synthesis, guided phoneme alignment, self-choir gated to vowels). No external music-generation services.

Chain #

bin/score.mjs (notation: 14 line entries, registers verses +0 / "arguing for" +7 / hooks +12) → bin/sing.mjs (per-line: ElevenLabs TTS → guided alignment → WORLD synthesis at harmony lock 0.875 (β = 0.125), f0 clamp 60–300 Hz, octave fit on, per-line QA: percentile conformance, click scan, whisper WER; then assembly → stems → consonant-duck mix over the bed → two-pass measured loudnorm) → out/blackboard.wav / .mp3 at −14.17 LUFS / −1.48 dBTP (target −14 / −1, SING_TARGET).

Chunked rendering: node bin/sing.mjs --prep 1 renders one uncached line and exits (resumable); a plain run reuses the cache and finishes the track. bin/texture.mjs builds the optional Ingold whisper stem and alt mix.

QA #

out/blackboard-sung-qa.json + per-line sidecars under out/sung/. All 14 lines pass conformance and the click scan (one advisory each on p3/b1). Whisper end-to-end WER (tracked, not a blocker — sung+choir audio is intrinsically whisper-hostile): 4/14 lines ≤ 0.25 strict; mix smoke transcript stays mostly legible. Best lines: v4 and b2 at WER 0.

Rights #

  • Lyrics are Ingold's spoken words (short lecture passage), sung by @jeffrey's own cloned voice — no lecture audio in the canonical master.
  • stems/ingold-texture-NONRELEASABLE.wav and out/blackboard-alt-texture-NONRELEASABLE.wav contain 4 s of the actual lecture recording ("when you stand at the blackboard…") as an intro whisper. Unlicensed source: these two files must never ship to DistroKid or any release channel.