Two ways to get a Zoom transcript
Zoom's own audio transcript comes with cloud recordings. It needs a Pro, Business, Education or Enterprise account, with cloud recording and audio transcription turned on in the account settings. After the meeting, Zoom can email the host when the recording is ready, and again when the transcript is. The transcript is a VTT file listed with the recording on the Recordings & Transcripts page of the Zoom web portal, where you can read it beside the playback, edit lines and download it. It is created in English unless you change the recording's language and have it transcribed again, and Zoom's help notes that you can edit the text to add capitalization and punctuation, which the transcript does not include.
Transcribing the recording here works for any Zoom recording you have as a file: a computer recording, which free and paid accounts can make in the Zoom desktop app, or a cloud recording you downloaded. Use it when Zoom made no transcript, or when you want to correct the text against the audio, label speakers and download it as a document, subtitles or a captioned video.
Find your Zoom recording
Computer recordings are saved by default in a Zoom folder inside your Documents folder, on Windows, Mac and Linux. After the meeting, Zoom converts the recording and opens the folder. Inside, the audio and video file is an MP4 named video plus a number, and the audio-only file is an M4A named audio plus a number. Recording to your computer works in the desktop app only, not in Zoom's mobile apps.
For a cloud recording, sign in to the Zoom web portal, open Recordings & Transcripts, choose the meeting and download its files.
Upload the M4A when you only need the words: it holds the meeting's sound without the picture, so it is a smaller upload. Upload the MP4 when you want to see captions over the video or save a captioned copy.
Transcribe it
- Open the video to text studio and choose the MP4 or M4A file.
- Pick the language people spoke, or let StemVocal detect it.
- To see who said what, choose how many people spoke, up to 8. Speaker labels use credits, and they are estimates: check short replies and moments when people talk at once.
- Read the transcript beside the playback, click any word to hear it, and rename each speaker label to the person's name.
A free job covers a recording up to 30 minutes, and you get 2 a day, shared with the other StemVocal tools. A longer meeting, up to 3 hours, uses one credit for each started 30 minutes, so an hour-long call uses two. For a series of recordings, choose several files to make a batch of up to 20.
When a project name or a colleague's name is misheard all the way through, Find & replace fixes every instance at once, and one Undo restores them.
Download what you need
| You need | Download |
|---|---|
| Notes or minutes | Word, with a timestamp and speaker at the start of each turn |
| Captions to upload with the recording | SRT or VTT |
| A captioned video to share | Video, with the captions drawn onto the picture |
| A list of turns to filter by speaker | CSV, with start and end seconds |
A captioned copy is made in your browser; add subtitles to video shows the size, style and position choices. A paid transcript can also be translated from English into 15 languages, or from any of them into English. Review machine translation before you share it.
Already have Zoom's transcript?
Zoom's transcript file is an ordinary VTT file, so you can open it in any text editor. If its captions run early or late by the same amount, the subtitle timing tool shifts every line at once. To read it in another language, the subtitle translator translates VTT files between English and 15 languages and keeps every line's timing.
Get a cleaner recording next time
Zoom's own advice for better transcripts applies here too: keep background noise to a minimum, ask people to speak clearly into their microphones without typing or side conversations, keep a microphone near the people who talk most, and use an external microphone rather than a built-in one. When people talk over each other, the transcript usually keeps one voice.
Sources: Zoom Support, "Using audio transcription for cloud recordings" (updated August 22, 2026), "Finding your computer and cloud recordings" (updated August 22, 2026), "Starting a computer recording" (updated October 13, 2025) and "Enabling or disabling computer recording". Checked October 11, 2026.
Good to know
Does Zoom transcribe meetings on a free account?
Zoom's audio transcript is for cloud recordings, which need a Pro, Business, Education or Enterprise account with cloud recording and audio transcription turned on. On a free Basic account you can still record to your computer and transcribe that file here.
Where are my Zoom recordings saved?
Computer recordings go to a Zoom folder inside Documents by default, on Windows, Mac and Linux. Cloud recordings are on the Recordings & Transcripts page of the Zoom web portal, where you can download them.
Which Zoom file should I upload?
The audio-only M4A file if you only need the words: it holds the meeting's sound and is a smaller upload than the MP4 video. Upload the MP4 when you also want captions over the picture or a captioned copy of the video.
Can it tell who spoke?
Yes, with speaker labels, which use credits. Choose the most voices you expect, up to 8, then rename each label. Labels are estimates and work best when people take clear turns.
Is it free?
Yes, for recordings up to 30 minutes. You get 2 free StemVocal jobs each UTC day without an account, shared with the other StemVocal tools. Longer meetings, up to 3 hours, use one credit per started 30 minutes.