Force-align an American English audio recording with its transcript at the word and phone level, producing a Praat TextGrid with word and phone tiers. The user selects the backend at invocation: `fave` (Montreal Forced Aligner with the english_us_arpa model — ARPA phone labels, heavy local install, pairs with FAVE-extract for formants) or `webmaus` (BAS WebMAUS API with eng-US — X-SAMPA phone labels, lightweight install, free for academic use, no API key). Use whenever the user asks to force-align an English recording, produce a word+phone TextGrid, or run MFA / WebMAUS on an interview.
