SRT and WebVTT conversion: timing is only part of the file
Understand timestamp separators, identifiers and metadata loss before downloading a converted subtitle.
The visible difference
A conventional SRT cue contains a numeric identifier, a timing line with comma milliseconds, and its display text. WebVTT begins with a WEBVTT header and uses period milliseconds. WebVTT cue identifiers are optional and can be words rather than consecutive numbers.
Changing the punctuation is enough for a simple example, but it is not a complete strategy for every file. WebVTT can contain comments, style blocks, regions, positioning settings and additional text markup. SRT does not carry an equivalent representation of every feature.
A minimal pair
SRT 1 00:00:01,250 --> 00:00:03,500 A clear example. WebVTT WEBVTT 1 00:00:01.250 --> 00:00:03.500 A clear example.
For this plain example, the timing and text can be retained while changing the wrapper and separators. The converter validates the source first and generates a separate output file. A malformed cue is not silently omitted to make the export appear successful.
Review what cannot be carried into SRT
When a WebVTT source contains header metadata, regions, styles or cue positioning, the interface asks you to acknowledge the limitations before an SRT export. The output uses consecutive numeric identifiers; WebVTT-only metadata and positioning are not transferred. Extended voice, class and ruby tags are removed from the display text.
Basic text tags and unusual player behavior can still differ. Test the resulting subtitle in the editor or player that will receive it. A successful browser conversion is not evidence that every downstream player preserves the same appearance.
Keep an audit trail
Save the original and the new file with different names. Export the JSON operation report if you need to record input and output formats, cue count and the metadata notice. That report contains process information, not a signed assurance of platform acceptance.
If the file contains unsupported structures or strange encoding, inspect it in a text editor before converting. The workbench expects UTF-8 subtitle text and does not read image-based subtitle tracks, recover deleted words or perform transcription.