When a video transcript helps
A transcript turns something you have to watch into something you can skim, search and quote. People use this tool to pull quotes out of recorded interviews, write blog posts from webinar recordings, create show notes for video podcasts, take notes from lectures, and keep a written record of recorded meetings.
If you need on-screen captions rather than a document, use the Subtitle Generator instead. It produces timed SRT and VTT files that video players and platforms such as YouTube can read.
How the transcription works
The tool uses Whisper, an open-source speech recognition model. By default it runs right in your browser: your device reads the audio track, detects the spoken language and turns the speech into text, with the time each sentence starts. The first time, your browser downloads the model once and keeps it for next time. The results appear in an editor where each line shows its timestamp.
On a phone or an older computer, you can choose to process the video on our server instead, which uploads the file and extracts its audio there. Either way, processing time depends on the video's length and is shown as a progress bar.
Supported formats and limits
- Video formats: MP4, MOV, M4V, WEBM, MKV, AVI. The video needs an audio track.
- File size and length: up to 500 MB and 60 minutes per video.
- Languages: auto-detect, or pick the spoken language from the list, which includes English, Spanish, French, German, Portuguese, Japanese, Korean, Chinese, Hindi and Arabic.
- Not available yet: speaker labels, translation, and importing a video from a link such as YouTube.
Tips for a more accurate transcript
- Clear speech close to the microphone gives the best results. Background music and crosstalk reduce accuracy.
- Select the spoken language if you know it, especially for short clips where auto-detection has little to go on.
- Check names, brand names and technical terms. These are the words speech recognition most often gets wrong.
- For a very large video file, convert it to audio first with MP4 to MP3 and upload the MP3 to Audio to Text. It uploads much faster.
Export options
Download the transcript as a plain-text .txt file, with or without [hh:mm:ss] timestamps at the start of each line, or copy it straight to your clipboard to paste into Google Docs, Word or Notion. Any corrections you make in the editor are included in the export.