SUBTITLES TO TEXT

SRT & VTT to plain text

Remove subtitle timestamps and export a TXT transcript. Choose cue paragraphs or joined text, with optional speaker labels.

Checking for a local draft…

From subtitles to readable text

  1. 1. Open your subtitle file

    SRT and VTT are supported. Keep the same file across tools.

  2. 2. Review the words

    Select a cue to edit its text. Save or cancel before exporting.

  3. 3. Download plain text

    TXT leaves out timing. Repeated lines and descriptions stay.

What this tool supports

TXT leaves out timestamps and formatting. Choose separate cue paragraphs or a joined transcript. Repeated speech and sound descriptions are kept. Speaker names can be retained or removed.

SRT or VTT · UTF-8 · Up to 2 MiB or 10,000 cues

CueRefine extracts text from an existing SRT or WebVTT file and downloads a UTF-8 TXT transcript. It removes cue timestamps and recognized formatting tags while keeping the words, repeated speech and sound descriptions.

A reproducible example

Synthetic sample. Download the input and open it in the tool above to reproduce this result. These examples are verified against the subtitle engine, not generated by a model.

Input · VTT
WEBVTT

00:00:01.200 --> 00:00:04.400
<v Alex>The result is <not final>.</v>

00:00:05.000 --> 00:00:06.000
<i>[Music]</i>
Output · TXT
Alex: The result is <not final>.

[Music]
Download example input (VTT)

Choose how the transcript reads

Cue paragraphs separate subtitle cues with blank lines. Joined transcript joins cues with spaces and flattens internal line breaks. Neither mode reconstructs sentences, summarizes speech or removes repeated caption fragments.

VTT voice annotations can become speaker labels, or you can turn that option off. Ordinary speaker names already written as dialogue are still text. Review the output before sharing.

What stays in the text?

Recognized subtitle formatting such as italic tags is removed. Literal text such as <not final> is preserved rather than silently discarded. Character entities are decoded after recognized tags are removed.

Sound cues such as [Music], repeated lines and ruby annotation text may still need manual editorial review. If you need timing later, keep the original subtitle file as well as the TXT export.

Supported files and processing

Free tools accept UTF-8 SRT and WebVTT, up to 2 MiB and 10,000 cues. Processing happens in your browser, without an account. The local draft is stored in this browser and can be cleared; it is not a cloud backup.

Questions about this tool

Can I turn audio or video into a transcript?

No. Start with an existing SRT or VTT file. This tool extracts subtitle text; it does not perform speech recognition.

Why are repeated words still in the transcript?

Rolling captions and repeated speech can produce repeated text across cues. CueRefine preserves it rather than guessing which words to delete. Review repetition manually.

Can I download TXT without saving an edit?

Save or cancel the active cue edit first. Export uses the saved document so the output matches the reviewed version.

Examples and product behavior checked on October 9, 2026.