Does the summary describe the picture?
No. Nothing is said about the images, the slides or the faces. The summary is a reading of the words that were spoken.
Video to summary
Summarize a video file
Summarize a video file: the picture is left alone, the sound track is read, and you get three sentences, up to five key points and the tasks.
3 files a day up to 30 minutes, free99+ languagesSpeaker separation and timestamps
In short
Drop a video on the widget and the picture is left alone: the engine reads the sound track, writes a transcript, and the summary is made from that text. Three sentences and up to five key points. The free plan is $0, three files a day.
How to get a summary of a video
- 1
Send the video file
mp4, webm and mov are accepted. A file may be up to 30 minutes; the free plan stops at 30 minutes and three files a day.
- 2
Wait for the transcript of the sound
The engine returns the words of the sound track, line by line, with the time of each line. The picture is never described and never analysed.
- 3
Ask the summarizer for the short version
Ask the plugin for Claude or ChatGPT: the tools return the retelling, the key points and the tasks that were mentioned. The plugin and the MCP server are part of Pro.
- 4
Read it against the video
Every line of the transcript carries the time it was said, so a sentence in the summary can be checked in a second.
What you get
- TXT
Plain text for an email, a note or a chat
- SRT
Subtitles for video editors and platforms
- VTT
Subtitles for a browser and a website
- CSV
A table: time, speaker, text
- DOCX
A Word document with speakers and times
- Markdown
Text with headings for notes
What comes out of a video
| What | Value |
|---|---|
| The input | a video file: mp4, webm, mov — the sound track is what gets read |
| The retelling | three sentences and up to five key points, taken from the transcript |
| The tasks | the sentences that name a task, with its owner and its deadline when they are there |
| Who asks for it | the plugin for Claude or ChatGPT, or the MCP server — both are part of Pro |
| One file | up to 30 minutes on the free plan, no minute limit on Pro; on the free plan 30 minutes and 3 files a day |
| The paid plan | $10 a month paid yearly, or $20 a month paid monthly |
The case
Only the sound is read
A video of a talk is usually a picture of a person and a long stretch of speech. The summary only cares about the second half of that, which is why the file can be a phone recording taken in a lecture hall. What you get back is a short account of what was said, and nothing about what was on the screen.
Tips
Split the recording before you upload it if it runs over 30 minutes.
The picture adds nothing to the summary; the sound carries it.
Check the timecode of a line before you quote it.
A video file eats the size limit faster than an audio one.
Questions this page answers
What if my video is longer than the limit?
One file may be up to 30 minutes. The free plan allows 30 minutes per file, so a long recording has to be split or moved to Pro.
Would an audio file give the same result?
As a rule, yes: the summary is made from the transcript, so the same transcript gives the same retelling. The transcript itself depends on the sound, so compare the first lines before you rely on it.
Is the English of my video processed elsewhere?
A pure English file may be sent to a third-party cloud speech provider in the United States, when that fallback is enabled on the server. A Russian or mixed file stays on our own engine.
Reading: 6 minutesUpdated 
Transcribsi editorial team
Drop a recording — the text arrives in a minute
Speakers by name, a time on every line, export to TXT, SRT, VTT, CSV and Markdown.
3 files a day up to 30 minutes, free99+ languagesSpeaker separation and timestamps