SRT to VTT Converter — Subtitles, Resync & Plain Text

Convert SRT to VTT, VTT to SRT, or strip the timings entirely and get a plain text transcript. Paste the subtitle text or pass file URLs in bulk. Cues that run early or late can be resynced by any offset in the same pass.

No account anywhere, no upload step, no API key. $0.005 per subtitle file converted — a file that produces no cues is recorded and not charged.

What it does

You have	You want	This does it
.srt from a transcription tool	.vtt for an HTML5 <track> or YouTube	outputFormat: "vtt"
.vtt from a browser player	.srt for Premiere, Resolve, VLC	outputFormat: "srt"
Either	A readable transcript for a description, blog post or summary	outputFormat: "text"
Subtitles 2.5s out of sync	The same file, corrected	shiftSeconds: -2

Input

{
  "subtitle": "1\n00:00:01,000 --> 00:00:04,000\nWelcome to the show.\n",
  "outputFormat": "vtt",
  "shiftSeconds": 0
}

Or in bulk, from URLs:

{
  "urls": [
    "https://example.com/episode-01.srt",
    "https://example.com/episode-02.srt"
  ],
  "outputFormat": "vtt"
}

Field	Type	Notes
subtitle	string	Raw .srt or .vtt contents. Source format is detected automatically.
urls	string[]	Direct links to subtitle files, up to 50 per run, 5 MB each.
outputFormat	vtt | srt | text	Defaults to vtt.
shiftSeconds	integer	−3600 to 3600. Negative if subtitles run late. Defaults to 0.

Output

One record per file:

{
  "ok": true,
  "source": { "type": "raw" },
  "sourceFormat": "srt",
  "outputFormat": "vtt",
  "cueCount": 2,
  "durationMs": 8000,
  "durationHms": "00:00:08",
  "shiftSeconds": 0,
  "characters": 118,
  "content": "WEBVTT\n\n00:00:01.000 --> 00:00:04.000\nWelcome to the show.\n\n…",
  "warnings": []
}

content is the finished subtitle file — save it straight to disk.

What it gets right

Subtitle conversion looks trivial and is full of small traps. This handles them:

• WebVTT hours are optional. 00:01.000 is one second in VTT, not one minute. Treating the first field as hours turns a one-second cue into a one-minute one; the parser decides by field count, never by position.
• Decimal separators differ. SRT uses 00:00:01,000, VTT uses 00:00:01.000. Both are read, and output is always written in the target format's convention.
• Short fractions. 00:00:01.5 means 500 ms, not 5 ms.
• VTT cue settings (align:start line:90%) after the end time are parsed and dropped rather than corrupting the timing.
• NOTE, STYLE and REGION blocks are metadata and never leak into subtitle text.
• CRLF line endings and a UTF-8 BOM are handled.
• Cue numbering is rebuilt from 1 on SRT output, so a shifted or filtered file is still valid.
• A negative shift clamps at zero instead of producing negative timestamps, and the cue count stays the same — which is what you are billed on and what you diff against.
• Plain text output strips <i> tags and {\an8} positioning, because those are presentation, not words.

Pricing

$0.005 per subtitle file converted. No start fee. A file that yields no cues — an HTML error page, a mislabelled download — produces an ok: false record explaining why and is not charged.

Limits

• 50 files per run, 5 MB per file.
• SubRip (.srt) and WebVTT (.vtt) in, the same two plus plain text out. SSA/ASS, TTML and SCC are not supported.
• Styling and positioning are preserved inside cue text on srt/vtt output, and removed on text output.

FAQ

Does it need an API key or an account? No. Paste text or pass a URL.

Can it fix subtitles that drift progressively? No — shiftSeconds applies one constant offset. Progressive drift is a frame-rate mismatch and needs rescaling, which this does not do.

Is my subtitle text stored? It goes to your own run's dataset on Apify, the same as any Actor output, and nowhere else.

Where do the subtitles come from? If you need to generate them first, see Audio & Video to Text (Whisper) — it outputs SRT, which this converts.

Changelog

• 0.1 (2026-09-10) — first release. SRT ↔ VTT, plain-text transcript, constant resync, bulk URLs. 30 local parser assertions cover the format traps listed above.

Who uses it

• Video developers converting SRT captions to WebVTT for an HTML5 <track> element
• Podcast and video editors shifting subtitles that run two seconds late
• Course creators turning caption files into a plain-text transcript for show notes
• Localization vendors converting a batch of VTT files back to SRT for an editing tool
• Accessibility teams preparing caption files in the format their player requires