Text Cleaner

Each switch does one job, and they always run in the same order: markup first (HTML, Markdown), then transcript clutter (timestamps, speaker labels), emojis and quotes, then spaces and lines (trim, collapse, duplicates, sort), then line breaks, empty lines and finally case. Next to every switch that is on, a count shows how many changes it made, so you can see at once if a switch did nothing or more than you expected.

It never rewrites, corrects or translates a word. It only removes spacing, line breaks, markup, timestamps, labels and emojis you chose, swaps quote marks, sorts lines or changes capital letters. The result is plain text in a box you can copy or download as a .txt file.

Text copied from a PDF keeps a line break at the end of every printed line, so a paragraph pastes as a column of short lines. Remove line breaks joins them with a single space: "The quote is valid" and "for 30 days." on two lines become "The quote is valid for 30 days."

Keep paragraph breaks is on by default: a blank line between paragraphs stays, so the paragraphs survive. If the PDF had no blank lines between paragraphs, everything joins into one; add an empty line where each paragraph starts and clean again. A word split with a hyphen across two lines keeps its hyphen, so check those by hand.

Paste a list with one entry per line (email addresses, leads, part numbers, names) and switch on Remove duplicate lines. The duplicate line remover keeps the first time each line appears and drops the repeats; spaces at the ends of a line do not count, and capital letters do not count either unless you tick Case-sensitive.

Blank lines are never treated as duplicates, so the spacing of a text stays. Sort A to Z or Z to A puts the list in order, with numbers in natural order (item 2 before item 10). The Clean up a list button does all of this in one go: trim, collapse spaces, remove empty lines, remove duplicates, sort A to Z.

Transcripts from meeting apps and caption files come with clutter. Remove timestamps takes out [00:01:23] and (00:01) anywhere, a timestamp at the start of a line such as 0:01:23.456, a time after a speaker name on its own line, and in caption files the WEBVTT header, the cue numbers and the timing lines with -->. A time of day inside a sentence, such as "we meet at 10:30", stays.

Remove speaker labels takes out a name or label of up to four capitalised words followed by a colon at the start of a line (Speaker 1:, SPEAKER_02:, Anna:, Dr. Lee:) and lines that hold only a label such as Speaker 3. Any capitalised word plus a colon at the start of a line counts, so a line such as "Note: call back" loses its label too: read the result before you send it. Repeated lines are kept by the transcript button, because "Yes." twice is real speech.