Features

View the transcript (only you)

What the transcript is

The transcript is the raw text of the video — taken from YouTube captions, or, when no captions exist, produced by AI transcribing the audio. It differs from "Section summaries": sections are AI-condensed, while the transcript is the word-for-word original.

How to view it

  1. Open any completed analysis.
  2. Scroll to the Transcript section near the bottom and click to expand (it loads on expand, so it never slows down the page).
  3. Each line has a timestamp — click it to jump to that moment on YouTube for easy line-by-line checking.

Privacy

Only you (the owner of the analysis) can see the transcript. Even if you share the analysis publicly, the transcript will not appear on the shared page — public pages only show summary-level content. This respects the video creator's content rights.

Note

The transcript is available once the analysis completes. When AI transcription is used (no captions), a few misheard words are normal; outlines and summaries are designed to be robust to such minor errors.

Names of people and brands

Speech recognition most often gets names wrong, replacing them with words that sound the same. When the transcript comes from AI transcription or auto-generated captions, we check the video's title, description and chapter names so that names in the outline, summaries, mind map, Q&A, quizzes and quotes follow the creator's spelling.

  • The transcript itself stays as it was recognized, so you may see a sound-alike word in the transcript while the analysis shows the correct name.
  • A name that the title, description and chapter names never mention can still be wrong. You can edit any text to correct it.

Who the people in the video are

While writing the outline and section summaries, we also check the video's title and description to work out who is speaking. That is why the chapters of an interview name the guest instead of just saying "the guest". This also applies to videos whose captions were uploaded by the creator: names mentioned in the title or description follow that spelling, and other names follow the captions.

  • In Chinese, "he" and "she" sound the same, so they cannot be told apart by ear. When the description says which one it is we follow it, but an interview may still occasionally get someone's gender wrong. You can edit the text to correct it.
  • People the title and description never mention can only be identified from what the video itself says.