Convert VTT to SRT
Upload or paste a WebVTT caption file, preview each cue, and download numbered SRT. This vtt to srt conversion does not need the original video.
Limits: 2 MiB UTF-8 and 10,000 cues. Conversion does not use the 1 GB transcription cap.
Conversion runs in the browser. Saving a signed-in task stores the subtitle text on the account. That save is not an offline-only promise. Conversion does not use transcription credits. The workflow is not fully local if you save.
How to convert a VTT file
Use the frame above. You do not open the speech workspace to finish this conversion.
- 1
Add the WebVTT text
Choose a .vtt file or paste the WEBVTT source. A file without that header is rejected.
- 2
Preview and correct cues
Edit start, end, and caption text. Warnings list NOTE, STYLE, REGION, cue settings, and tags that SRT cannot keep.
- 3
Download numbered SRT
Download works without an account. Sign in only to store the subtitle text in My records. Saving spends no transcription credits.
Practice with success.vtt. It includes a NOTE and STYLE block that must not appear in the SRT.
WEBVTT vs SRT compatibility
Both formats store caption text with start and end times. They do not share a header, a millisecond mark, or a styling model.
- File header
- WebVTT begins with WEBVTT. SRT has no header and numbers cues from 1.
- Milliseconds
- WebVTT uses a dot, such as 00:00:03.250. SRT uses a comma, such as 00:00:03,250.
- Hours
- WebVTT may omit hours, written as 00:01.000. SRT written here always includes hours.
- Styling
- WebVTT may include NOTE, STYLE, REGION, cue settings, and tags. SRT keeps timed text only.
A player that expects SRT will treat a WEBVTT header as junk. Convert when an editor asks for numbered SRT. The W3C WebVTT specification and the MDN WebVTT API guide describe the source format. Checked September 19, 2026.
If you do not have a VTT file yet, generate captions from audio on the SRT Subtitle Generator or the VTT Subtitle Generator. Those pages transcribe speech. This page maps an existing caption file.
What changes during conversion
Each usable cue becomes a numbered SRT block. Features SRT cannot represent are dropped and listed as warnings.
- NOTE, STYLE, and REGION blocks never become cue text.
- Cue settings after the arrow are dropped.
- Inline tags are stripped. Inner words stay.
- Empty cues disappear. Overlapping times stay, with a warning.
- WebVTT cue identifiers are discarded. Remaining cues are numbered from 1.
Keep the times, drop the stylesheet
Course teams often receive WebVTT from a web player and must hand numbered SRT to an editor. The millisecond values stay the same. 00:00:02.400 becomes 00:00:02,400. The original video is not read and is not required.
Illustration of a captioned course frame after conversion, not a customer result.
WhisperWeb vs in-browser drop-in converters vs server-upload converters
Checked September 19, 2026. Names below label the approach. This page does not link out to those products.
| Decision | WhisperWeb | In-browser drop-in | Server-upload converter |
|---|---|---|---|
| Where the file is parsed | WhisperWebAvailable: This page, in your browserCues appear in the tool frame before download. | In-browser drop-inLimited: Usually localPaste boxes often hide what they dropped. | Server-upload converterNot available: Uploaded firstHappy Scribe is the server-upload example, with no outbound link here. |
| Loss you can inspect | WhisperWebAvailable: Warnings on STYLE, NOTE, REGIONSkipped WebVTT features are listed beside the cues. | In-browser drop-inLimited: Often silentStyling can vanish without a note. | Server-upload converterLimited: Editor after uploadRestyling happens after the file is stored on a host. |
| Credits and related work | WhisperWebAvailable: No transcription creditsOptional My records stores subtitle text. Generate SRT from audio on the SRT generator if you lack a VTT. | In-browser drop-inAvailable: Usually unpaid pasteConfirm whether a paste is stored. | Server-upload converterNot available: Account workflowCaption desks often bill for transcription, not this mapping. |
Approach comparison, checked September 19, 2026. Happy Scribe is named only as a server-upload example.
Limits, privacy, and credits for VTT to SRT
This converter is a text mapping. Saving is not an offline-only promise.
The browser accepts up to 2 MiB of UTF-8 WebVTT and 10,000 cues. Those caps are shown in the tool frame. They are not the 1 GB media limit used for speech transcription.
Conversion does not upload media and does not spend transcription credits. A guest can convert and download. If you sign in and save, WhisperWeb stores the subtitle text on your account. The original video is never required.
Frequently asked questions
Will styling be preserved?
No. SRT has no STYLE block, REGION, cue settings, or inline tags. This converter strips that markup, lists a warning, and keeps the remaining words and times.
Do timestamps change?
The millisecond values stay the same within 1 ms. The written form changes from a dot to a comma, and hours are always printed. 00:01.200 in WebVTT becomes 00:00:01,200 in the download.
Do I need the original video?
No. Conversion reads the VTT text only. The video helps later when you check reading speed in a player, but it is not uploaded and not required to produce the SRT.