Every subtitle format does the same basic job: it pairs lines of text with start and end times. The differences are in what else the file can carry, such as styling, positioning and comments, and in which players and platforms accept it. Picking the right format saves a conversion later and keeps formatting you have already paid for.
This guide compares the formats you are most likely to meet, shows what each looks like, and explains which one to use for common destinations.
Subtitle formats at a glance
| Format | Extension | Timing written as | Styling and position | Typical use |
|---|---|---|---|---|
| SubRip | .srt | 00:01:02,500 | Minimal; some players read basic italic and bold tags | Video platforms, editing software, media players |
| WebVTT | .vtt | 00:01:02.500 | Cue position and alignment, CSS styling in browsers | HTML5 video on websites |
| Advanced SubStation Alpha | .ass, .ssa | 0:01:02.50 | Fonts, colors, outlines, exact positioning, effects | Styled releases, anime, karaoke |
| YouTube SubViewer | .sbv | 0:01:02.500 | None | Files downloaded from or prepared for YouTube |
| TTML / DFXP | .ttml, .dfxp, .xml | XML time attributes | Rich, defined by the profile | Broadcast and streaming delivery |
| SAMI | .smi | Milliseconds | HTML-like styling | Older Windows players |
| MicroDVD | .sub | Frame numbers | Limited | Legacy files tied to a frame rate |
SRT: the default for most uploads
SRT is plain text. Each cue has a number, a timecode line and one or two lines of text, with a blank line between cues:
1
00:00:01,000 --> 00:00:04,000
Welcome back to the channel.Almost every video platform, editor and player accepts SRT, which makes it the safest choice when you are not sure what a destination supports. The trade-off is that SRT has no standard way to set fonts, colors or position. Players apply their own style.
WebVTT: subtitles for the web
WebVTT is the W3C standard for subtitles and captions on HTML5 video. It looks close to SRT, with three visible differences: the file starts with a WEBVTT header, milliseconds follow a period instead of a comma, and cue numbers are optional.
WEBVTT
00:00:01.000 --> 00:00:04.000 line:85%
Welcome back to the channel.The setting after the timecode positions the cue on screen. Browsers can also style cues with CSS. If you run your own video player on a website, WebVTT is usually the format it expects.
ASS and SSA: when styling matters
Advanced SubStation Alpha files are scripts. They define named styles with fonts, sizes, colors and outlines, then list each line of dialogue against a style:
[Events]
Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
Dialogue: 0,0:00:01.00,0:00:04.00,Default,,0,0,0,,Welcome back to the channel.ASS can place text anywhere in the frame, show signs and on-screen translations, and animate karaoke lyrics. That power is also its risk: a tool that does not understand ASS styles can strip them when it converts or translates the file.
Other formats you may receive
- SBV is a simple format associated with YouTube. It has no cue numbers, and each start and end time sits on one line separated by a comma.
- TTML and DFXP are XML formats used in broadcast and streaming. Delivery specifications often name a specific profile, so follow the client's spec exactly.
- SAMI files look like HTML and appear mostly in older Windows workflows.
- MicroDVD files count frames instead of time. Converting them needs the right frame rate, or every cue drifts.
- LRC files time song lyrics line by line for music players.
Which subtitle format should you use?
- Uploading to YouTube, Vimeo or social platforms: SRT. It is widely accepted and keeps the file simple.
- Your own website player: WebVTT.
- Styled subtitles, signs or karaoke: ASS, and keep it as ASS through translation.
- Broadcast or streaming delivery: whatever the delivery specification names, often a TTML profile.
- Importing into an editor such as Premiere Pro or DaVinci Resolve: SRT is the most dependable choice.
Converting between formats
Converting toward a simpler format loses information. SRT to WebVTT is nearly lossless: add the header and change the millisecond separator. WebVTT to SRT drops cue positions and styling. ASS to SRT drops every style, position and effect, keeping only text and timing.
Before converting, ask whether the destination really needs a different format. When it does, keep the original file so you can return to it if the styling turns out to matter.
Working with subtitle formats in Aivently
Aivently accepts SRT, WebVTT, ASS/SSA, SBV, LRC, MicroDVD, TTML/DFXP, SAMI, MPL2, TMP, CSV and TXT files, and exports SRT, VTT, ASS, TXT and JSON. When you translate a file, cue identity, timing, tags and styles are preserved, and tested subtitle formats round-trip, so a styled ASS file comes back as ASS with its styles intact.
The source and the translation appear side by side in one editor, with warnings for overlaps, gaps, characters per line and reading speed. The file is measured first, so you see the exact credit total before a job runs.