How many characters per line in subtitles in 2026. Use about 32 to 42 characters, two lines, and split on a phrase so a phone can still read the cue.
On this page
Transcribe faster with File Transcribe
Upload audio or video, get speaker labels, timestamps, and editable text free to try.
How many characters per line in subtitles has a practical answer for most web video: about 32 to 42 characters, and two lines at most. Broadcast clients sometimes name a tighter number. A phone in a feed wants the short end of that range. A paragraph in one cue is already wrong.
This is a file check, not a font lecture. The rest of the pre-publish pass is the caption file checklist.
The number to use when nobody handed you a spec
Phone feed, LinkedIn, a Reel with a real caption track
- Line length to aim for
- About 32 characters
- Lines
- 2
YouTube and course video
- Line length to aim for
- About 37 to 42
- Lines
- 2
A client who cited Netflix, BBC, or a broadcaster
- Line length to aim for
- Their number, not yours
- Lines
- Whatever they wrote down
Netflix-style English guides and BBC subtitle guidance have both lived in this band for years, often at 42 or a bit under. They also care about reading speed, which a character count alone does not capture. If a contract names a spec, follow the contract. Confirm the current document on that client's site rather than a blog's memory of it.
W3C's caption overview is the accessibility framing, not a character law: captions and subtitles.
Why 42 keeps showing up
A full line of about 42 characters is roughly what an adult can read in the time a short cue stays on screen, on a television-sized picture, in one language, at a reasonable pace. It is a ceiling more than a target.
On a phone the picture is smaller and the cue fights the like button. Thirty-two characters leaves the line intact. Forty-two on a vertical video often wraps into a third line the player did not ask for. That third line is the failure, not the fact that you used 40 characters on a landscape YouTube video.
Count characters, then break on a phrase
Do not break because the counter hit 42 in the middle of a name.
Bad:
`We moved the billing address to the new` `office on Tuesday after the card failed`
Better:
`We moved the billing address` `to the new office on Tuesday`
The second cue can carry "after the card failed" if that is a separate beat. One idea per cue when you can. The closed captions vs subtitles distinction does not change this. Sound labels like `[applause]` still sit on their own line and still count.
Punctuation counts. Spaces count. A speaker name such as `ALEX:` counts. If you add names because two people are off camera, your budget shrinks. Say the name once, not on every line.
Two lines, then a new cue
One line is enough for a short reply. Two lines are the normal maximum for web captions. A third line covers the mouth, the lower third, or the LinkedIn controls.
If the editor shows a block of text, split it. The timing has to split too. A six-second paragraph that flashes as three lines fails even when each line is under 42. Reading speed is "can you finish it before it vanishes," not only "did the line wrap."
Fix that in the caption editor, not by shrinking the font in a burned-in export and hoping. Try the File Transcribe caption editor if the cues already exist. New captions start by uploading the video on the homepage and exporting SRT on Pro.
What changes the number
- Translation. A German or Finnish line is often longer than the English. Do not keep the English break and assume it still fits. Give the translation its own file.
- Kids' or language-learning video. Shorter lines. Slower cues.
- Shouty keyword captions in CapCut. Those are design, not this count. If you also need an SRT, follow the count in the file. See CapCut captions vs an SRT file.
- Burned-in text. The character guide still helps, but you are now also fitting a font size. Re-export if you change the words. You cannot patch pixels with an SRT.
LinkedIn's muted autoplay makes the short line matter more than on YouTube. The upload steps are in how to add captions to a LinkedIn video.
A 60-second check
- Open the SRT.
- Scan for a cue with three or more text lines.
- Scan for a line that runs past the low 40s, or past the low 30s if the video is vertical.
- Read the longest cue out loud in the time it stays on screen.
- Split on a phrase, export again, and spot-check the first minute on a phone.
Pro is about $19 a month and includes SRT and VTT. Free is about 50 minutes and exports TXT or PDF, not the caption file. Confirm on pricing.
FAQ
How many characters per line should subtitles be?
About 32 characters for a phone feed, and about 37 to 42 for YouTube or course video, with two lines maximum. Use the client's number when they gave you one.
Is 42 characters a hard rule?
No. It is the usual ceiling in English web and streaming practice. A 20-character line is fine. A 70-character line is not, unless you are failing the check on purpose.
Do spaces and punctuation count?
Yes. So does a speaker label at the start of the line.
What if my caption tool does not show a count?
Paste the longest line into any editor that shows a character count, or look at the SRT. If it looks like a paragraph, it is too long. The checklist version of this test is in the caption file checklist.
Should I use the same limit for SRT and VTT?
Yes. The limit is about reading, not about the extension. YouTube usually wants SRT. Some course players want VTT. The line length does not change with the extension.
Upload the video on the homepage, split the long cues, and export again before the file goes to YouTube or a client.
More guides
- Detect topics and keywords with AI
- AI sentiment and intent in transcriptions
- How AI transcriptions save time
- Test transcription accuracy
- Transcription guides
Further reading
Written by

Rasif Ali Khan
Founder, File Transcribe
I made File Transcribe to turn recordings into editable text without extra steps. I write these guides from the workflows I use myself, like meetings, podcasts, lectures, and the rest.
Related articles
All posts →Happy Scribe vs Otter vs File Transcribe for Subtitles (2026)
Happy Scribe vs Otter vs File Transcribe for subtitles in 2026. The deliverable is captions: Otter is the wrong tool, Happy Scribe is the studio, File Transcribe is the SRT you proof yourself.
How to Fix Subtitles for Accuracy (2026)
How to fix subtitles for accuracy in 2026. Proofread names, split lines, and resync timing from a transcript instead of fighting burned-in text.
File Transcribe vs Happy Scribe for Subtitles (2026)
File Transcribe vs Happy Scribe for subtitles in 2026. SRT/VTT you finish yourself versus a subtitle studio with burn-in and localization.
Closed Captions vs Subtitles, Explained (2026)
Closed captions vs subtitles explained simply. What each term actually means, when to use which, and how to generate either one from a video file.