Zeepit
How-to · 7 min read

How to transcribe a YouTube video into text, on phone or computer

Most public YouTube videos already carry an automatic transcript: open the transcript panel under the video and copy the text. When it is missing, unreadable, or too long to read, transcribe or summarise the video on your phone instead.

Does YouTube already give you the text?

Often, yes. YouTube generates automatic captions for most public videos, and you can read them as a transcript before installing anything.

On a computer: open the video, expand the description with ...more, then click Show transcript. A panel opens with the whole text, line by line, each line stamped with its time code. In the YouTube app on iPhone and Android, tap the description under the title and scroll down to the Transcript section.

Two things to check before you rely on it:

  • The button only appears when a caption track exists. Creators can switch captions off, and automatic captions are not produced for every language or every upload.
  • Automatic captions are raw: no punctuation, no paragraphs, no speaker names, and regular mistakes on proper nouns, brands and figures.

When a video carries subtitles written by a human, the transcript is far cleaner. You can usually tell within the first few lines.

How to copy a YouTube transcript without installing anything

For a short video, a copy-paste is enough.

  1. Open the transcript panel.
  2. Open the three-dot menu inside that panel and choose Toggle timestamps, so you are left with continuous text.
  3. Select everything and paste it into a notes app or a document.
  4. Reflow it: add paragraph breaks when the topic changes, restore punctuation, fix names.

On a phone the selection is fiddlier, because the panel scrolls along with the video. Copy in blocks rather than trying to grab an hour in one gesture.

Keep the timestamps if you plan to cite the video or come back to a passage; they are the fastest way to the right second. Strip them when you want something readable. On a ten-minute tutorial the whole thing takes two minutes. On a two-hour interview, the clean-up is the work.

Step by step

  1. Open the video and look for the transcript

    On a computer, expand the description and click Show transcript. In the YouTube app on iPhone or Android, tap the description and scroll to Transcript. The full text appears line by line.

  2. Copy the text and drop the timestamps

    Use the three-dot menu inside the panel to toggle timestamps off, select the text and paste it into a notes app. Then add punctuation and paragraph breaks.

  3. Paste the link for an instant summary

    No transcript, or a two-hour video you will not read in full? Paste the link into Zeepit on your phone and get a summary in short, medium or long.

  4. Share the audio or video file for the exact words

    For a word-for-word transcript, share the file with the app from Files or Drive. It uses only the audio track of mp4, mov, mkv and avi, up to four hours per recording.

  5. Check the names and numbers, then copy or export

    Re-listen to the passages you intend to quote and fix proper nouns and figures. One tap copies the text; exporting to PDF, Word or plain text is part of Zeepit Pro.

What to do when the automatic transcript is missing or unusable

Several situations leave you with nothing you can use:

  • The creator disabled captions, or the upload is recent and the automatic pass has not run yet.
  • The video is a live stream, mostly music, or recorded in a noisy room.
  • Several people talk over each other, or the accent is far from what the captions expect.
  • The spoken language was detected wrong and the text comes out as nonsense.

First, check the language selector at the top of the transcript panel: a video can carry several caption tracks, and the default is not always the right one. If the upload is only a few hours old, come back later.

If the text is still unusable, the audio has to be transcribed again — which means working from a file, or from the link, rather than from the caption track.

How to get a YouTube video's text from your phone

There is no upload page to open on a computer: everything happens on the phone, in two different ways depending on what you actually need.

A link, for the substance. Zeepit accepts YouTube links, podcast links and web page links, and turns them into an instant summary you can read short, medium or long. Paste the URL, wait, read. That is what you want when a 90-minute conference contains ten minutes of useful content.

A file, for the exact words. When you need the transcript word for word, feed the app the audio or the video file: pick it from Files or Google Drive inside the app, or tap Share on the file anywhere else and choose Zeepit (how the Share button works). It reads mp3, m4a, wav, ogg, opus, aac, flac and more, and it accepts mp4, mov, mkv and avi by using only the audio track. Up to four hours per recording.

That second route is for videos you own or already have the rights to: your own uploads, a webinar you recorded, a file a colleague sent you. Both work on iPhone and Android, the first conversions after install are free, and the usage is shown in the app.

Videos in another language: transcribe first, translate after

YouTube can machine-translate its captions on the fly. It is fine for following along, but the wording is usually too loose to quote or to hand to a client.

A cleaner sequence is to get the text in the original language first, check it, then translate. An error in the source transcript only gets bigger once translated.

Zeepit detects the spoken language on its own and produces the transcript, the summary and, when you need it, a translation, across dozens of languages — useful when the tutorial is in English and your notes have to be in Portuguese, Spanish, German, French, Italian or Turkish. See what the app does.

One more thing worth checking: some channels now publish several audio tracks. If the video settings offer one in your language, switch before opening the transcript panel.

What you are allowed to do with a YouTube transcript

Transcribing for yourself is ordinary use: notes for a class, a searchable version of a long conference, a readable text for someone who cannot follow the audio. There is nobody to ask.

Publishing is different. A full transcript posted online is a copy of someone else's work in another format, translated or not. As a rule of thumb, and rules differ from one country to another:

  • Quote short extracts, credit the channel, link to the video.
  • Ask the creator before republishing a full transcript or a translated version of it.
  • For your own channel, the transcript is yours, and YouTube Studio also lets an owner download the subtitle files.

Keep the source in mind as well: a transcript records what was said, not what is true. Check figures and claims before reusing them.

Frequently asked questions

Can I transcribe a YouTube video for free?

Yes. YouTube's own transcript panel is free on every video that has a caption track, and copying the text costs nothing. In Zeepit, the first conversions right after install are free, long recordings included, with no account and no card. The usage is shown in the app.

Why is there no Show transcript button on this video?

Because the video has no caption track. The creator may have turned captions off, the upload may still be processing, or automatic captions may not exist for that language. Live streams and music-heavy videos are frequent gaps too. In that case, the audio has to be transcribed itself.

How do I transcribe a YouTube video on iPhone or Android?

In the YouTube app, tap the description and scroll to the Transcript section. If it is missing or unreadable, paste the video link into Zeepit for an instant summary, or share the audio or video file with the app for the full transcript. No computer involved.

How accurate are YouTube's automatic captions?

Good on clear speech in a common language, weaker on strong accents, crosstalk, background music and technical jargon. They also arrive without punctuation or speaker names. Whatever tool you use, listen again to the passages you plan to quote: proper nouns, brands and numbers are where automatic text fails most.

Read next

← Back to the blog