WebVTT is a superset of SRT, so this conversion is lossy

The WebVTT format carries metadata and styling that SRT has no mechanism to express: optional cue identifiers, positioning instructions, and inline markup. Converting from VTT to SRT discards all of it, not because the tool is cutting corners but because the target format has nowhere to put it. Knowing exactly what disappears is the difference between a clean conversion and a surprise on someone else's player.

Exactly what this tool strips

• Cue identifiers — the optional label line before each timestamp. The parser recognises and steps over it, but never stores it.
• Positioning settings — align, position, line, size and vertical on the timestamp line. The parser captures them into a group it then never reads, so a caption pinned to the top of the frame to avoid covering on-screen text loses that placement entirely.
• STYLE blocks — recognised as legitimate WebVTT metadata and skipped without being counted as errors.
• REGION blocks — same treatment as STYLE: valid markup, cleanly discarded.
• NOTE comments — also skipped silently. These are authoring comments, not cues, so nothing is lost that a viewer would have seen.
• Inline markup — tags such as <v Speaker> or <c.classname> are not removed. The cue body is copied verbatim, so those tags survive into the SRT as literal text and will display on screen. If your VTT uses voice or class spans, strip them before or after converting.

That last point is the one that catches people out: everything else on this list vanishes, but inline markup passes straight through and becomes visible junk in the caption.

What does survive the conversion

The core data — timestamps and caption text, including multi-line text — comes through intact, and cues are renumbered sequentially from 1. Times are re-serialised with a comma before the milliseconds instead of a dot, matching SRT convention, and because they are held as integer milliseconds throughout there is no rounding drift. Blocks whose timestamps cannot be parsed are skipped and named in an amber warning; only a file with zero usable cues fails outright.

For the reverse conversion, see the SRT to VTT Converter, or work with YouTube's caption format via SBV to SRT.