The ElevenLabs subtitle generator turns speech in a video into a subtitle file. As of 21 October 2026, its subtitle generator page takes MP4, MOV or MKV files of up to 10 minutes or 50 MB, transcribes them with ElevenLabs' Scribe speech-to-text model, and lets you download SRT, VTT, TXT, DOCX, PDF or JSON. It covers 99 languages by that page's count, can translate the result, and labels speakers. There is a free version; a free ElevenLabs account includes 12 minutes of Scribe transcription a month.
What it gives you is a file, not a captioned video. To get captions drawn into the picture you either load the SRT into an editor, or build a video project in ElevenLabs Studio, which can burn styled captions into an export. Below is what each ElevenLabs tool does, what it costs, and when a tool that outputs a captioned MP4 is the simpler route.
What ElevenLabs' pages say (21 October 2026)
| Question | What ElevenLabs says | Where |
|---|---|---|
| What goes in? | MP4, MOV or MKV, up to 10 minutes or 50 MB | Subtitle and caption generator pages |
| What comes out? | SRT, VTT, TXT, DOCX, PDF or JSON; the caption page adds HTML | Subtitle and caption generator pages |
| Languages | 99 on the subtitle page; 90+ on the caption page and in the docs | Product pages and docs |
| Is there a free version? | Yes; paid plans raise the limits | Caption generator FAQ |
| Free transcription allowance | 12 minutes of Scribe a month on the Free plan | Speech to Text pricing table |
| Translation | Yes; a separate subtitle translator covers 150+ languages | Subtitle generator FAQ, subtitle translator |
| Speaker labels | Yes, up to 32 speakers | Caption generator page, docs |
| Burned-in captions | In Studio video projects, styled captions are burned into the export | Studio guide |
| Longer files | Up to 3 GB and 10 hours through Speech to Text | Docs |
The ElevenLabs subtitle tools, one by one
ElevenLabs has several pages for the same engine, which is why the search results look confusing.
Subtitle generator and caption generator. Both landing pages run on Scribe and take short clips. The caption generator is pitched at short-form video: it detects the spoken language, tags non-speech sounds such as laughter, and says its SRT and VTT files import into CapCut and Premiere Pro, where you style them.
Speech to Text in your account. This is where longer work happens. The developer docs list audio and video input, files up to 3 GB and recordings up to 10 hours, with word-level timestamps and up to 32 speakers. Transcripts export as plain text, JSON, HTML, SRT or VTT.
The subtitle editor. ElevenLabs' subtitles guide describes a subtitling mode inside the transcript editor. You edit text in place, drag handles on a timeline to change a cue's start and end, and split, merge or add cues. The editor checks three rules as you work, characters per line, lines on screen and cue duration, and shows red or green feedback; you can change the rules from the panel's menu. Export is SRT or VTT. The guide also mentions a human subtitling service from $2.20 a minute, in beta for select users.
Subtitle translator. A separate page takes SRT, VTT, SUB, ASS, STL or TXT files you already have and translates them into 150+ languages, exporting the same range of formats.
Studio. ElevenLabs Studio, its project editor, has video projects with a captions panel. Its guide says a Style tab sets font, color, size and placement, those settings are burned into the exported video, and the canvas can be 16:9, 9:16, 4:5 or 1:1.
Free and paid limits
The landing pages say you can create captions for free and that paid plans raise the limits, but they do not give a free quota. The speech-to-text page does, for Scribe used in the ElevenLabs app:
| Plan | Price | Scribe transcription a month |
|---|---|---|
| Free | $0 | 12 minutes |
| Starter | $6 a month | 1 hour 31 minutes |
| Creator | $22 a month, $11 for the first month | 6 hours 7 minutes |
| Pro | $99 a month | 30 hours 18 minutes |
ElevenLabs' main pricing page describes plans in credits rather than minutes: 10,000 credits a month on Free, with Starter at $6 a month or $5 billed yearly. If you also use its voice tools, check how much of your plan transcription will use. In Studio, the guide says video exports on the Free plan carry a watermark, and paid plans export without one.
Styling: what you can and cannot do
The subtitle generator page mentions customizing fonts, colors and layouts. In practice, the files themselves carry no style: an SRT or VTT holds text and timings, and the look comes from the player or editor that shows it. YouTube, for example, draws its own captions over an uploaded SRT.
Studio is the ElevenLabs route to a styled, burned-in result. Its guide covers font, color, size, placement and a caption template. It does not describe word-by-word highlighting or animated caption presets of the kind short-form creators often use, so test a project before you plan a series around it.
When you need burned-in styled captions instead
If the goal is a vertical clip for Reels, Shorts or TikTok with the words animated on screen, a file is one step short. You would export the SRT, then style and burn it in another editor. The burn-in guide for HandBrake, VLC and FFmpeg covers that route if you already have a file.
CapsAI skips the file. Its AI subtitle generator takes the MP4 or MOV itself, up to 10 minutes, transcribes the speech in one of 11 spoken languages, and lets you fix every word and timing in the browser before it renders one MP4 with the captions burned in. There are 37 caption styles built for short-form video, 8 free and 29 included with any pack, plus AI emoji sync on the Creator and Pro packs. Pricing is one-time: $4 for 20 minutes, $9 for 60 and $19 for 180 (₹49, ₹99 and ₹299 in India; €4, €9 and €19 in the euro area). There are 3 free minutes to start, and watermark-free export comes with every paid pack. The burn subtitles into video page explains the output.
The 11 languages are English and 10 Indian languages, including Hinglish written in Roman script. That suits creators who post in those languages; anyone working in Spanish, French or Japanese needs a different tool.
When ElevenLabs is the better pick
Be clear about what CapsAI does not do. It has no SRT, VTT or TXT download, no subtitle-file import, and no translation. Captions come out in the language spoken, inside the video only. ElevenLabs is the better choice when:
- You need a subtitle file for YouTube, a video player, a course platform or a client
- You need captions translated into another language
- Your language is outside CapsAI's 11
- The recording is longer than 10 minutes, or audio only
- You want subtitle rules such as characters per line enforced as you edit
- You need an API or live subtitles for events
If you already have subtitle files from ElevenLabs or anywhere else, CapsAI's free browser tools can convert and adjust them, such as the SRT to VTT converter. Those tools run in your browser, never upload the file, and are separate from the captioner, which does not read subtitle files.
For a wider comparison of file tools and burn-in tools, including ElevenLabs, see the best AI subtitle generators compared.
How to get an SRT from ElevenLabs
- Sign in and open Speech to Text, or use the subtitle generator page for a clip under 10 minutes.
- Upload the video and turn on the option to include subtitles.
- Open the transcript, switch to the subtitle view, and check the rule warnings for line length and cue duration.
- Fix names, numbers and any cue the editor marks in red.
- Export SRT or VTT, then upload it to YouTube or your player, or load it into your editor to style.
What to do next
If you need an SRT or VTT, or a translation, use ElevenLabs: the 12 free Scribe minutes a month are enough to judge the transcript on your own audio. If you need a captioned vertical clip ready to post, upload the video to CapsAI and use the 3 free minutes to see the draft in a few styles before you buy a pack.
