Transcript Preprocessor

Input Text

Trint formatted transcript Removes timecodes and formats speaker names (e.g., "[00:08:03] Cindy Weinstein And so..." becomes "Cindy Weinstein: And so...")

Carry forward speaker name to paragraphs without a speaker

Preprocessing Options

Remove timecodes (e.g., " 01:08:" or " 1:08:") Convert literal \n\n to paragraph breaks Remove spaces before punctuation (e.g., "hello ." → "hello.") Convert \" to ' (escaped double quotes to single quotes) Ensure space after ellipsis (...) Remove double spaces

Remove Page Headers (Optional) Enter text that page headers begin with. Leave empty to skip header removal.

Transcript Preprocessor

Input Text

Preprocessing Options

Output