Observed arrival · 2026-09-29
RhetoricMap makes debate transcripts legible, sentence by sentence
A free, in-browser analyzer labels transcript sentences as attacks, policy arguments, both, or neither, and shows the cues behind each label.
- For
- Debate researchers, students, and transcript analysts
- Worth noticing
- Its name-matching cue accepts user-supplied opponent names and suppresses a match when the name appears in the speaker’s own name.
Field notes
The analyzer expects speaker names at the start of transcript lines and offers a field for comma-separated opponent names. Its page says the classifier scores sentences against visible cues, with user-entered names added at runtime; it also describes escaping metacharacters and sorting names longest-first. Rather than forcing every sentence into an attack or policy category, it retains both-at-once and neither labels for mixed or procedural remarks.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
One card from the complete issue