Skip to content

docs: preserve jokes/puns + homophone-flatten footgun - #5

Merged
evnchn merged 1 commit into
mainfrom
docs/preserve-jokes-homophone-footgun
Jul 18, 2026
Merged

docs: preserve jokes/puns + homophone-flatten footgun#5
evnchn merged 1 commit into
mainfrom
docs/preserve-jokes-homophone-footgun

Conversation

@evnchn

@evnchn evnchn commented Jul 18, 2026

Copy link
Copy Markdown
Contributor

TL;DR

Comedy/skit inputs live or die on the pun, and the skill said nothing about preserving jokes. Worse, homophone puns are invisible to the ASR — Whisper "corrects" the rare pun to the common word, so the gag never reaches the transcript. This adds a step-3 clause + one footgun row. Diff is +2/−1 lines.

Verdict: learned from a real localization this session; the pun-flattening was confirmed against the video's own burned-in caption (not inferred). Chinese + non-redundancy checked by a fresh lineage (Codex).

What changed

  • Step 3 (Translate): Preserve the joke, not just the words — trans-create an English pun that carries the same mechanism; the subtitle is the joke's primary carrier (the dub rides over the ducked original).
  • New footgun row: ASR silently flattens a homophone pun — the burned-in caption spells out the literal pun character, so OCR is usually the only reliable recovery path. Bridged to (and contrasted with) the adjacent gag-caption row so the two don't read as contradictory.
The real case that earned this

A Mandarin office-revenge skit (youtube.com/shorts/wh2_aBn_3MQ). Two homophone puns bookend it, and Whisper flattened both:

On-screen caption (truth) Whisper transcript Meaning
人工 人工 AI (智能) → "hemorrhoid-tech" (痔能) — the whole premise
"shelve the project" (搁置 gēzhì) → "hemorrhoid removal" (割痔疮 gēzhìchuāng) — the finale

and are both zhì; 搁置割痔 exactly in sound, with appended. Neither pun surfaced in ASR; both were recovered by OCR'ing the caption band. The English localization rendered them as "Artificial Intelligence" → "Artificial PAIN-telligence" and the finale as "sit on it" (shelve) → hemorrhoids so the joke survives with zero Chinese knowledge.

Fresh-lineage review (Codex, gpt-5.x)
  • Chinese all accurate (/ both zhì; 人工智能=AI; 人工痔能=valid character-substitution pun; 搁置=shelve; 割痔疮=hemorrhoid removal).
  • New row is distinct, not redundant with the gag-caption row.
  • Flagged "the ONLY recovery path" as overreach (visuals/reactions can also reveal a pun) → softened to "usually the only reliable recovery path," which also removes a self-contradiction with the row's own "Tell" hint.

Disjoint from open PR #4; branches off main.

Comedy skits live or die on the pun, and Whisper silently normalizes a
homophone pun to the common word — so the gag never reaches the
transcript and no proper-noun fix recovers it. Add:
- step 3: preserve the joke (trans-create puns; the subtitle is the
  joke's primary carrier, since the dub rides over the ducked original).
- a footgun row: the burned-in caption spells out the pun character, so
  OCR is usually the only reliable recovery path.

Learned localizing a real Mandarin skit: 人工智能 (AI) -> 人工痔能 and
搁置 (shelve) -> 割痔疮 (hemorrhoid removal), both flattened by ASR,
both recovered from the on-screen caption.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
@evnchn
evnchn marked this pull request as ready for review July 18, 2026 19:43
@evnchn
evnchn merged commit 0782292 into main Jul 18, 2026
@evnchn
evnchn deleted the docs/preserve-jokes-homophone-footgun branch July 18, 2026 19:43
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant