Key points
Knowledge tree
Couldn’t load the mind-map viewer. Check your connection and reload the page.
Drag to move, pinch or Ctrl + scroll to zoom, click a dot to collapse or expand a branch.
Turn any video into text. Free.
Accurate AI transcription that runs entirely in your browser — your videos are never uploaded. Transcribe one file or a whole batch, then get an AI summary and mind map. No sign-up, no watermark.
Transcribing…
Keep this tab open. Processing happens on your device, so speed depends on your computer.
Transcript ready
Get an AI-written summary, the key points and a knowledge tree (mind map) of this transcript.
This sends the transcript text — not your video — to our server, which passes it to Anthropic’s Claude AI to write the summary. We don’t save it.
Writing the summary… this usually takes 10–30 seconds.
How it works
-
1
Pick your files
Choose one video or audio file, or many at once. They stay on your device — nothing is uploaded.
-
2
AI transcribes it
OpenAI's Whisper speech-recognition model runs locally in your browser and detects the language automatically.
-
3
Download & summarize
Get TXT, SRT and VTT files — one by one or all in a ZIP — plus an optional AI summary with a knowledge-tree mind map.
Frequently asked questions
Is it really free?
Yes. Because the transcription runs on your own computer, there are no server costs to pass on to you. AI summaries are free too, with a fair-use limit of 20 per hour. No account, no trial, no watermark.
Are my files uploaded anywhere?
No. Your video is decoded and transcribed inside your browser tab, and the speech-recognition model is downloaded from Hugging Face and cached so later visits start faster. If you click “Generate summary”, only the transcript text is sent to our server, which passes it to Anthropic’s Claude AI to write the summary. We don’t save it.
Can I transcribe several files at once?
Yes. Select or drop as many files as you like. They are transcribed one after another, and when they’re done you can open each one or download everything as a single ZIP.
What is the summary and knowledge tree?
With one click, AI reads your transcript and writes a short overview, the key points, and a knowledge tree — a mind map that organizes the main topics and ideas. You can explore the mind map in the page, download it as an interactive web page, or download everything as Markdown, which mind-map apps such as XMind can import.
Which languages are supported?
Whisper understands 99 languages, including English, Chinese, Spanish, Hindi, Arabic, French, German, Japanese, Korean, Portuguese and Russian. Leave the language on “Auto-detect” or pick it yourself for best accuracy.
Why is the first run slower?
The first time you use a quality level, your browser downloads the model (about 40–250 MB depending on the quality and your device). After that it is loaded from your browser cache.
How long can my video be?
There is no fixed limit. Long recordings simply take longer, and very large files (over about 1 GB) may exceed your browser's memory. A recent laptop with a modern browser such as Chrome or Edge gives the fastest results.
Which export formats are available?
Plain text (TXT) and two subtitle formats with timestamps: SRT, supported by YouTube, Premiere Pro, DaVinci Resolve and most players, and WebVTT (VTT) for the web.