How to Transcribe a Zoom Recording (Free and Paid Options)

This guide is for anyone who hosted or attended a Zoom meeting, has the recording, and needs the words on a page. By the end you will know where Zoom saved the files, whether Zoom's own transcript covers what you need, and how to turn the audio into an editable transcript with speaker names that exports as a Word document or as captions.

Where Zoom puts your recording

Zoom records in one of two places, and that decides your options.

Record on this computer saves the meeting locally. When the meeting ends, Zoom converts the recording and writes the files into a Zoom folder inside your Documents folder. You can change that location in the Zoom desktop app under Settings, then Recording. Each meeting gets its own dated subfolder with a video file in MP4, an audio-only file in M4A, and a chat log if anyone typed in the chat. The exact file names vary by version; older releases used names like zoom_0.mp4 and audio_only.m4a, and current ones use longer names with a numeric ID. Either way, the file you want ends in .m4a.

Record to the cloud sends the recording to Zoom's servers instead. Open the Zoom web portal in a browser, sign in, and click Recordings in the sidebar (some accounts show it as Recordings & Transcripts). Each cloud recording lists its files, usually a video MP4 with the shared screen or speaker view, an audio-only M4A, and sometimes a transcript. Click Download to save them. Cloud recording is only offered on paid Zoom plans, and the host decides whether other participants may download.

If you were a participant rather than the host, ask the host for the audio-only file; it is the smallest thing to send.

What Zoom's own transcript gives you

Zoom can generate an audio transcript of a cloud recording. The host's account needs a paid Zoom plan with cloud recording, and the Audio transcript option must be switched on in the cloud recording settings before the meeting is recorded. If it was on, the transcript appears next to the recording in the web portal, where you can read it alongside the video or download it as a VTT file. Speaker labels come from the participant names, which works well when everyone joins from their own device and badly when several people share one laptop in a room.

The quality is reasonable for clear English with one person speaking at a time, and weaker with accents, overlapping talk and specialist vocabulary. You can correct lines in the portal, but it is slower than editing a document, and getting the text into Word or plain text means copying it out yourself. Check your account settings for which languages the transcript supports.

Two things are often mistaken for a saved transcript:

• Live captions during the meeting are generated on the fly for the people in the call. In current versions the host can allow participants to save the caption text, which gives a plain text file, and in many setups the captions are simply gone when the meeting ends. Either way, this is not the cloud transcript.
• Local recordings never get a Zoom transcript. If you chose Record on this computer, you need a separate tool.

So Zoom's transcript is the right answer when you have a paid account, recorded to the cloud with the setting on, everyone spoke clearly from their own device, and reading it in the portal is enough. In every other case, carry on.

Upload the audio to RocketScribe

RocketScribe accepts both the MP4 and the M4A. Pick the M4A. It holds exactly the same audio track as the video at a fraction of the size. Do not convert the M4A to MP3 or WAV first; the original file always gives the best result.

1. Open the Zoom recording transcription page or the home page and drop in the M4A. The file is sent in parts, so if the connection drops, Retry continues where it stopped. The M4A to text page does the same job if you got the audio file from someone else.
2. Choose the spoken language, or leave it on auto-detect. Set it yourself when the first minute is small talk in a different language from the rest of the meeting. Meetings that switch between Chinese and English are handled.
3. Turn on Identify speakers. This is the setting that makes a meeting transcript readable, because it separates the lines by voice instead of giving you one long block of text. If you know how many people spoke, enter the number; a four-person call with the count set to four comes out cleaner than one left to guess.
4. Start the transcription.

Speaker identification is a paid feature, included on paid plans and with Minute Pack minutes. On the Free plan you still get the full transcript with timestamps, just without the voice separation.

Fix the speaker names and anything misheard

The transcript arrives as timestamped paragraphs with a player underneath. Click any line to jump the audio to that spot. Speakers come back as Speaker 1, Speaker 2 and so on, because the system groups voices by sound and has no idea who anyone is. Click a label, type the person's name, and every line from that voice is renamed at once. If one person was split across two labels, merge them.

Names, product terms and acronyms are the usual mistakes in meeting transcripts. Correct them in place; the text is editable directly, and paragraphs can be split or merged if the breaks landed in the wrong spot.

Export for minutes or captions

What you export depends on what the transcript is for.

• DOCX for meeting minutes or a record to circulate. Speaker names and timestamps can each be switched on or off; most people keep the names and drop the timestamps.
• SRT or VTT if you want captions on the video. Upload the MP4 to your video platform together with the SRT. If captions are the only thing you need, the MP4 to SRT page starts from the video file.
• AI prompt copies the transcript with a ready-made prompt, so you can ask ChatGPT or Claude for a summary and action items.

What to expect from meeting audio

Zoom audio is already compressed, and people on laptop microphones in echoey rooms are harder to transcribe than someone on a headset. Two people talking over each other will produce broken lines with any tool, and a group sharing one laptop will often be grouped as a single speaker. If a particular meeting matters, ask people to use headsets and to let each other finish. If you would rather keep the audio on your own machine, running Whisper locally works as well, though you give up the editor and the speaker labels and have to handle the exports yourself.

Try it on your own recording

Find the M4A in your Zoom folder or the web portal and upload it on the Zoom recording transcription page. If you only have the video, MP4 to text takes the file as it is.