Turn DFXP subtitle files into SRT subtitles quickly—upload your DFXP and download the converted SRT when it’s ready.
DFXP (Distribution Format Exchange Profile) is the original W3C profile of TTML — a verbose XML format that broadcasters and streaming services use to carry captions along with styling, colors, and on-screen positioning. SRT (SubRip) is the opposite: a numbered, plain-text list of timed cues that practically every player and editor opens without a plugin. This converter reads each timed paragraph out of the DFXP XML and rewrites it as SRT, so the file becomes something you can actually edit, upload, or drop next to a video. It runs entirely in your browser — the file is parsed on your own device and never uploaded to a server.
| Property | Value |
|---|---|
| Full name | Distribution Format Exchange Profile |
| Relationship | The original profile of TTML (W3C Timed Text Markup Language) |
| First W3C draft | November 2004, as "Timed Text (TT) Authoring Format 1.0 – DFXP" |
| Underlying standard | TTML 1.0, a W3C Recommendation (November 2010); TTML1 defines the DFXP Transformation, Presentation, and Full profiles |
| Structure | XML: a <tt> root wrapping <body> / <div> / <p begin="…" end="…"> cues |
| Timing | Clock time (HH:MM:SS.mmm), plus frame- and tick-based forms |
| Styling | tts:* attributes and named regions — color, font, alignment, position |
| Extensions | .dfxp, .ttml, .xml |
| Used by | Adobe Flash captioning (FLVPlayback component), Netflix, HBO, SMPTE-TT, broadcast pipelines |
| Property | Value |
|---|---|
| Full name | SubRip Subtitle |
| Origin | The SubRip DVD-ripping tool (early 2000s); a de-facto standard, never formally ratified |
| Structure | Sequentially numbered cues separated by blank lines |
| Timing | HH:MM:SS,mmm with a comma decimal, written as start --> end |
| Styling | Essentially none; some players honor HTML-like <i> and <b> tags |
| Extension | .srt |
| Native support | VLC, MPV, mpv, YouTube, subtitle editors, smart TVs — nearly everything |
| Best for | Universal playback, quick editing, translation, and sidecar caption files |
.dfxp, .ttml, or .xml file, or paste the XML straight into the input panel. The cues appear immediately, and because the tool runs in your browser, the file never leaves your device.<p begin="…" end="…"> paragraph and rewrites the clock time from DFXP's HH:MM:SS.mmm to SRT's comma-separated HH:MM:SS,mmm, numbering the cues 1, 2, 3 in order.<br/> tags become SRT line wraps, and the surrounding XML styling tags are stripped so you are left with clean plain-text cues.| Element in the DFXP | Result in the SRT |
|---|---|
| Cue text | Preserved verbatim, including Unicode |
| Cue timing (clock time) | Preserved, converted from .mmm to ,mmm |
<br/> line breaks |
Preserved as newlines inside the cue |
| Cue numbering | Generated fresh, sequentially |
Colors, fonts, tts:* styling |
Dropped — SRT has no equivalent |
| Regions and positioning | Dropped — every cue becomes a plain centered block |
<head> metadata, xml:lang, profile declarations |
Dropped |
SRT is a text-and-timing format by design, so the trade is deliberate: you gain universal compatibility and lose the presentation layer. If the styling matters, keep the original DFXP or edit the SRT afterward in a dedicated tool such as Subtitle Edit or Aegisub, which model color and position directly.
Almost. DFXP is the profile name TTML shipped under before the "TTML" label was settled — W3C's early-2000s drafts were literally titled "Distribution Format Exchange Profile," and the term survives in profile designators today. A .dfxp file is TTML XML with a specific profile, so a well-formed DFXP and a well-formed TTML look nearly identical inside: a <tt> root, a <body>, and <p> cues with begin and end attributes. This converter treats .dfxp, .ttml, and .xml the same way.
No — and that is expected. SRT has no concept of regions, color, font, or on-screen position, so this converter reads only the dialogue text and the cue timing out of the XML and discards the presentation layer. In our testing, a Netflix-style DFXP with region-positioned, colored cues produced an SRT that keeps every line of dialogue and every timecode exactly, while the region and tts:color attributes are gone from the output. If you need to retain styling, keep the original DFXP or rebuild it in a subtitle editor.
The DFXP almost certainly uses frame-based (HH:MM:SS:FF) or tick-based (500t) timing rather than plain clock time. This converter reads clock-time begin/end values in the HH:MM:SS.mmm form; frame counts and tick rates depend on a declared frame rate or ttp:tickRate that a lightweight in-browser parser cannot resolve reliably. If your cues collapse to zero, re-export the DFXP from its source with clock-time stamps, or run it through a full subtitle editor first.
Yes. Text is handled as UTF-8 throughout, so accents, Cyrillic, Greek, Arabic, and CJK scripts survive the rewrite unchanged. If a downstream player shows garbled characters, that player is probably expecting a legacy code page rather than UTF-8; re-saving the .srt in the encoding its documentation names usually fixes it, since the conversion itself does not alter the characters.
Three main places. Broadcast and post-production tools export DFXP/TTML because it is the W3C interchange format for captions. Streaming services — Netflix and HBO among them — deliver captions in TTML profiles, so ripped or exported caption tracks often arrive as .dfxp or .xml. And older Adobe Flash video used DFXP timed text through the FLVPlayback captioning component, so archived Flash projects tend to carry .dfxp sidecars.
Yes — SRT to DFXP reverses the trip, wrapping each SRT cue back into TTML <p begin="…" end="…"> paragraphs. Because SRT carries no styling, the round trip cannot recreate colors or positioning that the original DFXP had; it only restores the XML container and timing. Keeping an untouched copy of the original DFXP is the only way to preserve its full presentation.
No. This is a client-side converter: the DFXP is parsed and rewritten entirely in your browser, and the file never leaves your device. That is also why it is instant — there is no upload, queue, or server roundtrip, even for a feature-length caption track, and you can keep converting after the page has loaded even with the network disconnected.
Use DFXP to VTT, which produces the WebVTT captions HTML5 <track> elements expect. It works the same way — text and timing carry over, styling is dropped — so pick VTT when your destination is a web player and SRT when it is a desktop player, editor, or media-server sidecar. For any other target, the DFXP converter hub lists the full set of output formats.