How to transcribe a meeting or lecture recording
A Zoom file goes to Video to Text, a voice recorder file to Text Transcription. An hour costs about 44 tokens plus a few for the Essence with action items.
Updated November 2, 20262 min read
The goal is a text you can search, quote and send around, plus a short summary with decisions and tasks. Which tool to use depends on what kind of file you have.
Which tool to use
| What you have | Tool | Why |
|---|---|---|
| Zoom, Teams or Google Meet recording (MP4), a webinar video | Video to Text | videos up to 3 GB; you can close the app while it works |
| A lecture on YouTube | Video to Text with Paste link | up to 2 GB; nothing is downloaded to the phone |
| A file from the phone's voice recorder (M4A, MP3) | Text Transcription | audio files up to 100 MB |
| A lecture that is happening right now | the phone's voice recorder, then Text Transcription | the recording stays on the phone if something goes wrong |
Both tools detect the language and split the text by speaker. Neither has a start button or a language choice: work begins as soon as the file is added.
Example: an hour-long Zoom call
Open the Tools tab and tap Video to Text.
Tap Choose and pick the Zoom recording.
Keep the app open while it shows Compressing video... and Uploading video....
When Recognizing speech... appears, close the app. A push notification comes when the text is ready; if it cannot be delivered, an email does.
On the Text tab read the transcript by speaker with timecodes. Tap a timecode to hear that moment.
On the Essence tab check Key points: for a meeting it often includes Questions and answers and Action items. Chapters shows the parts of the call.
Tap Export to PDF to send the summary to colleagues, or Discuss in chat under the transcript and ask for a follow-up email.
How much it costs
Recognition costs about 0.73 tokens per minute: an hour-long call is about 44 tokens. The Essence adds a few tokens. Tokens are taken after the transcript is ready, so a failed attempt costs nothing. See token prices for each tool.
If the audio file is larger than 100 MB
Long recordings in WAV or high-quality M4A often go over 100 MB, and the app answers "Please choose a file no larger than 100 MB." Options:
Convert the file to MP3. An hour of MP3 at 128 kbps is about 58 MB.
If the recording is a video, add the video file to Video to Text: it accepts up to 3 GB.
If the recording is online, add it by link: up to 2 GB.
How to get a better transcript
Put the phone closer to the person who talks most, or to the middle of the table.
For a call, use the recording made by Zoom or Teams: it sounds cleaner than a phone held to the laptop speaker.
Speakers are labeled Speaker 1, Speaker 2 and so on. The app does not know their names and cannot rename them; replace the labels after you copy the text.
Contact support
Describe the problem and we will reply by email. Add the file name or job ID if you have one.