Audio Transcriptor
Start

Turn subtitles into plain text

Get the readable text out of an .srt: no numbers, no timestamps and no forty-character line breaks.

  1. 1File
  2. 2Conversion
  3. 3Download

What this tool does

A subtitle file is not a text: it's a text chopped up to fit under a video. Every couple of seconds the sentence breaks, and a paragraph ends up spread across five numbered blocks. Fine for watching, useless for reading, quoting or feeding into anything else.

This puts it back together. Sentences are joined where the subtitle cut them, and a new paragraph starts where there's a real break — a silence of over two seconds, or a line that ended in a full stop. You can keep the timestamps if you're using it to navigate the recording, or drop them if what you want is prose.

What we handle that others don't

  • Paragraphs rebuilt by silences and full stops, not by dumping every subtitle on its own line.
  • Numbering and timings removed, or kept as a readable [1:23] if you want to find the moment again.
  • Formatting tags stripped: no <i> left in the middle of a sentence.
  • Speaker names kept when the file has them.

How it works

  1. 1

    Drop the .srt

    Or paste it. VTT files work here too.

  2. 2

    Choose what goes in

    With or without timestamps; line by line or joined into paragraphs.

  3. 3

    Copy or download

    Straight into your notes, your document or wherever it's going.

In detail

Why subtitle text reads so badly

A subtitle isn't cut where the sentence ends, it's cut where it stops fitting. The rule almost everyone follows is two lines of around forty characters, held on screen long enough to read, and that splits sentences in half with no regard for punctuation.

That's why an SRT opened in a text editor reads in fits and starts: loose numbers, timestamps in between, sentences broken every second line. The work here isn't stripping the numbers, which is the easy part — it's putting back together what was split, so what's left reads straight through.

What plain text is good for

Searching inside it, quoting a passage in a document, handing it to someone who isn't going to open a player, and feeding it to a language model for a summary. In all of those, timestamps get in the way more than they help.

Where they do help is when reviewing: if the text exists to send you back to the video to check something, keeping a time in front of each paragraph turns the document into an index. That's why it's a switch, not a decision we make for you.

Questions

And if what you have is the video?

Upload it and you'll get the transcript with timestamps, the speakers separated by name and a summary with the key points. You can start without an account.

Transcribe my recordingSee prices

Related tools

Worth reading