About this tool
Auto-transcription is a miracle with a rough edge: the text arrives littered with ums, uhs, you knows and likes, chopped into arbitrary fragments wherever the speaker paused, and formatted in whatever casing the recognizer felt like. Cleaning that by hand takes longer than the recording. This cleaner does it in one pass. Paste raw transcript text and it removes a comprehensive list of English filler words and false starts, merges fragment lines into flowing sentences, restores sentence capitalization, normalizes stray whitespace, and optionally breaks the result into one paragraph per sentence for maximum readability — ideal for show notes, articles derived from podcasts, or feeding clean text to an LLM. A statistics line shows how many fillers were removed and how the word count changed, so you can see the surgery was conservative. The original meaning is never altered: only disfluencies and formatting are touched. Everything runs locally in your browser, so confidential interviews and unreleased episodes stay private.
How to use Transcript Cleaner
- Paste the raw transcript into the input area.
- Toggle options: filler removal, line merging, per-sentence paragraphs.
- Press Clean transcript and review the result and statistics.
- Copy the cleaned text or download it as a .txt file.
Frequently asked questions
Which filler words are removed?
Common English disfluencies: um, uh, er, ah, like, you know, I mean, sort of, kind of, basically, actually and similar, matched as whole words.
Will it change the meaning of my text?
No. Only fillers, whitespace and casing are touched — never word order or content words.
Does it work for non-English transcripts?
The filler list is English-focused; other languages still benefit from line merging and whitespace normalization.
Is my transcript uploaded?
No. All processing runs locally in your browser, safe for confidential interviews.