kennethreitz a62dce399a CI: pre-download NLTK data for g2p-en
Same pre-warm as the Dockerfile — without the tagger the g2p fallback
dies and the OOV-word tests fail.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-07 01:53:15 -04:00

RhymePad

A scratchpad for poets & rappers. Write in the pad; rhymes are detected phonetically (CMU Pronouncing Dictionary) and color-coded as you type — end rhymes, internal rhymes, and slant rhymes included. Comes with a rhyme/synonym lookup, draggable stanza reordering, and synthesized beats with adjustable tempo.

Run

$ uv run uvicorn app:app --reload

Then open http://127.0.0.1:8000.

How rhyme detection works

  • Every word is mapped to its CMU phonemes (pronouncing), with lyric-friendly fallbacks (runnin'running, possessives, and a spelling heuristic for made-up words).
  • Perfect rhymes share everything from the last stressed vowel on (tonight / light / flightAY T). These are matched anywhere in a line — that's the internal rhyme detection.
  • Slant rhymes share just the vowel sounds from the last stressed vowel (hold / coal). Applied to line endings that didn't find a perfect match.
  • Same color = same sound. Underline = line-ending rhyme (the stanza's a/b/a/b scheme); soft glow only = internal rhyme.

API

  • POST /api/analyze {"text": "..."} → token spans, rhyme groups, per-stanza schemes
  • GET /api/lookup?word=light&mode=rhyme|near → frequency-ranked rhymes / near rhymes, grouped by syllable count

Synonyms come from the free Datamuse API, client-side.

S
Description
No description provided
Readme ISC 351 KiB
Languages
Python 60.2%
HTML 39.4%
Dockerfile 0.4%