Convert VTT to TTML Online

How to Convert VTT to TTML Online

1. Upload Your VTT File: Drag and drop your .vtt file or click "Add Files" to select. You can also paste WebVTT text directly. Batch upload is supported, and conversion runs in your browser — files never leave your device.
2. Confirm Target Format: TTML is preselected as the output. Each <p> element in the resulting TTML is mapped from a WebVTT cue, with timing attributes (begin / end) and <span> styling derived from cue settings and class tags.
3. Review the Output (Optional): Open the converted file in any text or XML editor to add a TTML profile (e.g., IMSC 1.1, EBU-TT-D, SMPTE-TT), regions, or stylesheet references that downstream packagers require.
4. Convert and Download: Click "Convert". Files process in your browser session — no sign-up, no watermark, no upload to a server.

Why Convert VTT to TTML?

WebVTT is the de facto subtitle format for HTML5 video and HLS web playback, but most broadcast and OTT delivery specs ask for TTML — an XML-based, profile-driven standard published by the W3C. Converting VTT to TTML is what turns a working web caption file into a deliverable asset for streaming services, archive workflows, and packaging pipelines that won't accept plain-text VTT.

• Netflix delivery — Netflix's post-production specs require timed text to be delivered as TTML1 with a .xml or .ttml extension; Japanese-language assets must conform to the Netflix IMSC 1.1 Text Profile. A VTT track captured from a web player won't pass QC.
• Amazon Prime Video, Disney+, and other OTT — Prime Video's content guide accepts SRT, WebVTT, or TTML in UTF-8, but TTML is the only format that survives transformation to broadcast head-ends without re-authoring.
• fMP4 / DASH / HLS packaging — Unified Streaming, AWS Elemental, and other packagers wrap TTML in fMP4 using the stpp codec (ISO/IEC 14496-30), versus wvtt for WebVTT. Many CMAF workflows expect TTML on input even when the player output is WebVTT.
• Authoring in DaVinci Resolve, Adobe Premiere, or EZTitles — Pro tools import VTT but export TTML profiles for finishing. Converting first lets you keep timing and text while you re-style inside the editor.
• Multi-profile distribution from one master — TTML can be transformed into EBU-TT-D for European broadcast, SMPTE-TT for US captioning, or back to WebVTT for the web. A single TTML master replaces several format-specific files.
• Long-term archive — TTML is XML, schema-validated, and namespaced. It survives toolchain changes far better than plain-text WebVTT, which has no formal schema.

WebVTT vs TTML — Format Comparison

Property	WebVTT (.vtt)	TTML (.ttml / .xml)
Structure	Plain text with WEBVTT header and cue blocks	XML document with <tt> root and namespaces
Standard	W3C WebVTT (Candidate Recommendation Draft)	W3C TTML2 (REC, 2018); profiles include IMSC 1.2, EBU-TT-D, SMPTE-TT
Timecode	00:00:00.000 --> 00:00:02.000 delimiter line	begin="00:00:00.000" end="00:00:02.000" attributes
Styling	Inline cue settings + optional CSS via STYLE blocks	Attribute styles, referenced <style> elements, regions
Positioning	Cue settings: position, line, size, align	Dedicated <region> elements with percentage extents
Profiles	Single specification, no profiles	DFXP, SMPTE-TT, EBU-TT-D, SDP-US, CFF-TT, IMSC 1/1.1/1.2
Primary use	HTML5 <track>, HLS WebVTT segments	OTT delivery (Netflix, Disney+), broadcast captioning, archive
fMP4 codec	wvtt	stpp (ISO/IEC 14496-30)

TTML Profile Quick Guide

Profile	When to pick it	Notes
Plain TTML1	Generic interchange, Netflix non-Japanese delivery	Default output; valid for most authoring tools
IMSC 1.1 Text Profile	Netflix Japanese delivery, modern OTT	Subset of TTML2; requires ttp:profile declaration
EBU-TT-D	European broadcasters (BBC, ARD, France TV), HbbTV	Distribution profile; mandates UTF-8 and specific region rules
SMPTE-TT	US broadcast captioning, FCC-compliant workflows	Adds SMPTE-specific metadata and image-based caption support
DFXP	Legacy Flash / Adobe / older packagers	Original TTML1 distribution profile; still accepted by many tools

xconvert exports a generic TTML1 document. If your spec requires a stricter profile, open the file in a text editor and add the appropriate ttp:profile attribute and region definitions — the timing and text payload are profile-portable.

Frequently Asked Questions

Will styling from my WebVTT file carry over to TTML?

Basic timing and cue text always carry over. Inline cue settings (position, line, align) and class spans map to TTML attributes and <span> elements per the W3C TTML-WebVTT mapping, but TTML supports a richer set of style and region constructs than WebVTT does. Expect to add regions, fonts, and colour palettes inside your finishing tool if your delivery spec requires them — the converter gives you a structurally valid TTML skeleton to build on.

Is the output valid TTML1 or TTML2?

xconvert emits TTML1-compatible XML with the standard http://www.w3.org/ns/ttml namespace, which validates against both TTML1 and TTML2 parsers. It does not declare an IMSC, EBU-TT-D, or SMPTE-TT profile by default — you can add ttp:profile="http://www.w3.org/ns/ttml/profile/imsc1.1/text" (or the relevant profile URI) at the top of the document if your downstream tool requires it.

Why does Netflix ask for .xml or .ttml and not .vtt?

Netflix's localization pipeline does QC, language tagging, and forced-narrative handling on TTML because the format carries explicit metadata and structured regions that WebVTT cannot represent unambiguously. The post-production spec page is explicit: "TTML1 with .xml or .ttml extension" for all subtitle and SDH files, with IMSC 1.1 required for Japanese.

Can I convert VTT to TTML offline or in bulk?

Yes. xconvert runs the conversion entirely in your browser — uploaded VTT bytes never reach a server, and the page works after first load with the network disabled. For automation, the open-source sandflow/ttconv CLI is a good fit for CI pipelines and supports WebVTT, TTML, SRT, SCC, and STL.

My VTT has chapter cues and metadata. What happens to them?

WebVTT supports three cue kinds — subtitles, chapters, and metadata. Standard VTT→TTML mapping treats only subtitle cues as timed text; chapter and metadata cues are typically dropped because TTML has no direct equivalent. If you need chapters preserved, export them separately (for example as an MP4 chapter atom or a sidecar JSON) before converting.

How does TTML handle multiple languages compared to VTT?

A WebVTT file represents one language track; multi-language web playback uses one .vtt per language. TTML can carry multiple languages in a single document using xml:lang attributes on <div> or <p> elements, which is one reason it is preferred for live broadcast distribution where switchable language tracks are encoded together.

Will line breaks and positioning survive the conversion?

Soft line breaks inside cue text become <br/> elements in TTML. Positioning encoded in VTT cue settings (line:70%, position:50%, size:80%) maps to region or tts:origin / tts:extent style attributes. Exact pixel-accurate placement may need a manual pass in your authoring tool, because TTML positioning is profile-dependent — IMSC, EBU-TT-D, and SMPTE-TT define their own conventions.

Can I go back to WebVTT after converting?

Yes — use TTML to VTT for the reverse direction. The W3C mapping is lossy in both directions: TTML→VTT drops region styling and most attribute styles, while VTT→TTML produces a minimal skeleton without profile-specific regions. If you need to round-trip without loss, keep the original VTT alongside the TTML master.

Looking for other subtitle conversions?

xconvert also supports VTT to SRT for plain SubRip output, VTT to DFXP for the legacy TTML1 distribution profile, VTT to ASS for Aegisub-style advanced substation, and SRT to TTML if you're starting from a SubRip source.