By the Recapo.ai Editorial Team · Fact-checked July 10, 2026
If you want to know how to add closed captions on YouTube, the short answer is that everything happens inside YouTube Studio, and you have three native routes: upload a caption file (like an SRT), paste a transcript and let YouTube auto-sync the timing, or type the captions by hand. What the official docs gloss over is the step that actually decides whether your captions are any good — getting an accurate, timeline-aligned SRT before you ever open Studio, and cleaning up YouTube's auto-generated track when you'd rather edit than rebuild. This guide covers both: the exact clicks inside Studio, and the upstream work that separates captions that read cleanly from a wall of mistimed guesses.
Closed captions vs. subtitles vs. burned-in: get the terms right
Before the clicks, get the vocabulary straight, because the word you search decides the workflow you land in.
Closed captions (CC) are a separate text track stored alongside your video — usually an SRT or VTT file. Viewers toggle them on and off with the CC button. Because the text is a real data layer and not part of the picture, YouTube can index it for search, offer auto-translation, and let each viewer choose whether to see it. This is what "add closed captions on YouTube" almost always means, and it's the accessible, discoverable option.
Open captions (also called burned-in or hardcoded) are pixels baked into the frame in your editor before upload. They're always on and can't be toggled, translated, or searched — but they survive re-uploads and look identical everywhere. Short-form creators lean on burned-in captions for Shorts, Reels, and TikTok; long-form YouTube uploads are better served by proper closed captions.
Strictly speaking, "subtitles" translate dialogue for a hearing viewer, while "captions" transcribe all meaningful audio (including speaker cues and sound) for deaf and hard-of-hearing viewers. YouTube uses the two terms loosely and treats YouTube subtitles and captions through the same Studio panel, so don't overthink the distinction — just know which layer you're producing.
| Closed captions (CC) | Separate text track (SRT/VTT) | Yes, via CC button | YouTube Studio |
| Open / burned-in | Text baked into the frame | No, always on | Your editor, before upload |
| Auto-captions | ASR track YouTube generates | Yes, and editable | YouTube Studio |
If you also publish to other platforms and want the burned-in version, our guide on how to add subtitles to a video walks through that side.
The three native ways to add closed captions on YouTube
Every method starts the same way. In YouTube Studio, open Subtitles from the left menu (or open the video, then Subtitles), pick the video's original language, and click Add next to the language you want. YouTube then offers three ways to create the track.
| Upload file | You supply a timed caption file (SRT) | As good as your file | You want full control, or already have a transcript |
| Auto-sync | You paste plain transcript text; YouTube times it | Solid on clear speech, drifts on hard audio | You have a script but no timings |
| Type manually | You type lines in the editor and set timings | Slow but precise | Short video, no transcript on hand |
There's also a passive fourth track: automatic captions, which YouTube generates by speech recognition on most videos without you doing anything. They appear labeled "(automatic)" and are editable — more on fixing those below.
For anything longer than a minute or two, Upload file is the method to reach for. It hands you total control over wording, line breaks, and timing, and it's the only route where you can proofread every line before YouTube ever sees it. The catch — the part the docs skip — is that you need a good SRT to upload. That's the real work, so let's do it properly.
Why an uploaded SRT beats auto-sync and auto-captions
Auto-sync and auto-captions are convenient, and on clean, single-speaker audio they can be close to right. But both hand control to a black box, and both fail in the same predictable places.
An uploaded SRT solves all three because you set the words, the timings, and the breaks. You proofread once, upload once, and the result is exactly what you approved — the difference between captions that disappear into the video and captions viewers notice for the wrong reasons. So the question becomes: how do you build that SRT without hand-typing every timestamp?
How to create an accurate, timeline-aligned SRT before you upload
This is the step the official documentation waves past — and it's the one that decides your caption quality. The goal is a clean SRT with the right words and honest per-line timings, produced in minutes rather than typed by hand. The path is transcribe, edit, export.
- Transcribe the audio to text. Start from the video's audio, not from memory. A speech-to-text tool can produce a transcript draft to read and correct instead of retyping from the timeline; treat the output as unapproved until it is checked against the recording.
- Fix the words that recognition fumbles. Scan for the four things transcribers miss most — names, homophones ("their/there"), numbers (prices, dates), and jargon — and correct them while you have the text in front of you. This is faster now than after the caption timings are locked.
- Generate timed, line-broken captions. Turn the corrected transcript into a properly segmented caption track with an auto-subtitle-generator or an editable ai-subtitle-generator, so each line carries a real in/out timestamp and breaks at sensible points instead of mid-phrase.
-
Export as SRT. Export the track as an
.srtfile. SRT is the most widely supported caption format for uploads and the simplest to open in a text editor if you need a last-minute tweak — YouTube accepts several formats, but SRT is the safe default.
The output of this loop is the exact file YouTube's Upload file option wants, already proofread. You've moved the accuracy decision upstream, where you control it, instead of hoping the auto tools guess right.
Step by step: upload an SRT to YouTube Studio
With a clean .srt in hand, adding it is a two-minute job. Here's the full path to upload SRT to YouTube through YouTube Studio captions.
- Open YouTube Studio and click Subtitles in the left sidebar (or open the specific video, then choose Subtitles).
- Confirm the video's original language. If you haven't set it, YouTube asks — pick the language actually spoken, because it governs how the track is labeled.
- Click Add next to the language you're captioning, then choose Upload file.
-
Select "With timing" when prompted (your SRT already contains timestamps), and pick your
.srtfile. - Review in the editor. YouTube shows your captions beside the video. Scrub a few spots to confirm lines land on the right words and nothing drifts.
- Click Publish. The captions go live; viewers can now toggle them with the CC button, and YouTube can offer auto-translation on top of your accurate base track.
If you're captioning in more than one language, repeat from step 3 with each additional SRT. Because you built the source track carefully, every translation YouTube generates inherits your correct names and timing rather than an auto-caption's guesses.
How to edit YouTube's auto captions
Sometimes you don't want to rebuild — you want to fix the free track YouTube already made. You can edit YouTube auto captions directly in Studio, which is worth doing on older uploads or when the audio was clean enough that the machine got most of it right.
- In Subtitles, find the language row marked (automatic) and open it.
- YouTube loads the auto-generated text in the editor. Work down it and correct the predictable misses — names, homophones, numbers, and any garbled jargon — while nudging timings that landed early or late.
- Publish to replace the raw automatic track with your corrected version.
Editing in place is quickest when the auto track is mostly right. When it's badly off — heavy background music, crosstalk, a strong accent it couldn't follow — you'll spend less time transcribing clean and uploading a fresh SRT than fighting a bad automatic track line by line. If auto captions are a recurring pain point, how to improve auto-caption accuracy breaks down the input-side fixes that cut errors at the root.
SRT gotchas and troubleshooting
A few small formatting issues cause most "my captions won't show" and "the timing is off" complaints. Check these before you blame YouTube.
None of these need a support ticket; they're all decided by the file you upload, which is one more reason to build the SRT deliberately rather than accept whatever the auto tools produce.
Where Recapo fits
Recapo is a browser-based AI video workspace — nothing to install — and it lives upstream of the YouTube Studio steps above. It transcribes your audio, gives you a word-level editable transcript, lets you fix names, numbers, and jargon beside the video, generates properly line-broken caption timings, and exports a subtitle file you upload to YouTube. It accepts MP4, MOV, and other common formats, up to 6GB total per task — enough for a long upload you're captioning after the fact — and the same project can also resize to 9:16, add a voiceover, or build a cover if you're repurposing the video into Shorts.
The practical fit: Recapo doesn't post captions to YouTube for you — the Add > Upload file > Publish clicks still happen in Studio, and that part is free. What it does is make the file those clicks are waiting for: an accurate, timeline-aligned SRT you didn't have to hand-type, with the names and timing already correct. If you caption one video a month, YouTube's own tools may be all you need. If you're captioning at volume, in multiple languages, or repurposing long uploads into short clips, doing the transcribe-edit-export loop in one tab is the job it's built for. Plans are on the pricing page.
FAQ
How do I add closed captions on a YouTube video?
Do it in YouTube Studio. Open Subtitles, pick the video's language, click Add, and choose one of three methods: Upload file (an SRT you prepared), Auto-sync (paste a transcript and let YouTube time it), or Type manually. For anything past a minute or two, uploading a proofread SRT gives the cleanest result because you control the exact words, line breaks, and timing before YouTube ever sees them.
What file format does YouTube use for captions?
YouTube accepts several caption formats, but SRT (.srt) is the most widely supported and the easiest to work with. It's a plain-text file listing each caption's number, start and end timestamps, and text, so you can open and tweak it in any text editor. Save it as UTF-8 to avoid garbled accented or non-Latin characters, and it will upload cleanly through the Upload file option.
Can I edit YouTube's automatic captions instead of uploading my own?
Yes. In Subtitles, open the language row labeled "(automatic)," correct the text and timing in the editor, and click Publish to replace the raw machine version. Editing in place is fastest when the automatic track is mostly right. If the audio was noisy or had overlapping speakers and the auto captions are badly off, it's usually quicker to transcribe cleanly and upload a fresh SRT than to fix it line by line.
Should I use closed captions or burned-in subtitles on YouTube?
For standard YouTube uploads, use closed captions — the toggleable CC track that YouTube can index for search and auto-translate. Burned-in (open) captions make sense when you're repurposing the same video into Shorts, Reels, or TikTok, where hardcoded text keeps captions visible in muted, autoplay feeds. Many creators do both: a proper CC track on the long upload, and a burned-in version on the short clips cut from it.
Why are my uploaded captions not showing up or out of sync?
Two common causes. If they don't appear at all, you likely saved a draft instead of clicking Publish, or uploaded a plain transcript as a timed file — re-upload as "With timing" with a real SRT. If they drift, the SRT and the video are misaligned, usually because the video was trimmed after the captions were timed or the frame rates differ. Rebuild the SRT from the exact audio you uploaded and don't re-trim the video afterward.
Ready to caption your next upload without hand-typing a single timestamp? Create a free account, upload your video — MP4 or MOV, up to 6GB total per task — transcribe it to a word-level editable transcript, fix the names and numbers beside the video, and export a clean, line-broken SRT. Then it's a two-minute job in YouTube Studio: Subtitles → Add → Upload file → Publish. You keep the free part free; Recapo just makes the accurate file those clicks are waiting for, so your captions read like you wrote them — because you did.
References and official sources
Originally published at Recapo.ai.
Top comments (0)