Both transcripts (plain text or timed segments) are split into words, normalised...
https://ton.halowerk.com/v1/transcript-compareOverview
Both transcripts (plain text or timed segments) are split into words, normalised per options (NFKC, case folding, punctuation removed) and aligned with a word-level Levenshtein alignment. Consecutive non-matching words form one difference region with the text of A, the text of B, kind (replaced, only_in_a, only_in_b), word positions, times (from segment times, interpolated by character position), surrounding context and flags for numbers and pure formatting differences.