A transcript is the source, in text
Video is slow to search and awkward to quote. Text is neither. A transcript turns an hour of talking into something you can skim, search with your eyes, copy and cite.
SumAI gives you the full transcript of any video it summarizes, with a timestamp on every line.
What people use transcripts for
- Quoting accurately. Copy the exact sentence instead of paraphrasing from memory.
- Checking a summary. Read the passage behind a key point to see whether the nuance survived.
- Taking notes. Paste the relevant section into your notes app and trim it down.
- Reading instead of watching. Some people take in text faster than speech, especially in a second language.
- Finding a moment. Use the timestamps to return to the right part of the video.
How YouTube’s own transcript works
YouTube has a transcript too. According to YouTube Help, you open a video’s description and select Show transcript, and the option is offered for videos that have captions.
It is worth knowing two things about it on a phone. First, it lives inside the YouTube app’s video page, so it is built for following along rather than for taking text elsewhere. Second, it depends on the same thing every transcript does: a caption track.
SumAI’s transcript sits next to the summary, key moments and Ask AI, and it is made to be copied.
Which videos have transcripts
A transcript exists when a video has captions. There are two kinds:
- Creator captions, written or uploaded by the channel.
- Automatic captions, generated by YouTube’s speech recognition. YouTube Help lists the supported languages and notes that quality varies with accents, background noise and audio quality.
YouTube also explains why automatic captions are sometimes missing: the video is still being processed, the language is not supported, the audio is poor, several people talk at once, or the video begins with a long silence.
If a video has no captions of either kind, there is no transcript to show, in SumAI or anywhere else.
Transcript, summary and key moments together
The three views answer different questions:
| View | Answers |
|---|---|
| Summary | What is this video about? |
| Key moments | Where is each topic? |
| Transcript | What exactly was said? |
Most of the time you read the summary. When something matters, you jump to the key moment. When you need the exact words, you open the transcript.
Last updated November 12, 2026.