Zoom Transcription: Native vs Third-Party in 2026

Zoom does this natively on Pro+. Here's when that's enough — and the path that works when it isn't.

Zoom transcribes meetings natively on Pro tier and above($16.99/mo+), with AI Companion summaries included in the same plans. Free hosts, non-hosts who only have a local MP4, and anyone whose meeting is in one of the 80+ languages Zoom doesn’t support need a third-party transcription tool. DeluxeScribe handles Zoom MP4 and M4A files in 99 languages with speaker labels, 60 minutes free. Below: a 2026 feature matrix by Zoom tier, an honest decision tree, the non-host workflow nobody covers, and the retention reality that bites researchers and lawyers later.
  • 60 minutes free
  • No credit card
  • 99 languages
  • Speaker labels

Last verified June 26, 2026

TL;DR — pick your path

Zoom transcription is several different things depending on who you are and what tier you’re on. Pick the row that matches your situation.

Your situationBest pathCost
Host on Pro+, English/one of 19 supported languagesZoom native (enable Audio Transcript)Included with $16.99+/mo plan
Host on Free / BasicLocal recording → upload MP4 to a serviceFree tier → ~$10/mo
Host on Pro+ but need a language Zoom doesn’t coverCloud-record + download MP4 → upload to multi-language service~$10/mo
NOT the host — host won’t shareLocal recording your side OR personal note-taking bot → uploadFree tier → ~$10/mo
Need transcript past your org’s recording retentionDownload .vtt now, OR re-transcribe from your own MP4~$10/mo if re-transcribing
Regulated (HIPAA, attorney-client)Zoom Pro+ with active BAA, or self-hosted WhisperPlan cost + BAA setup

What Zoom natively does (by tier, 2026)

Zoom’s transcription features are layered across tiers and feature flags. The summary below reflects 2026 Zoom pricing and KB documentation; verify on zoom.us/pricing before relying on it.

TierCloud recordingCloud transcript (VTT)AI Companion summaryBAA available
Free / BasicNo (local only, 40-min cap)NoNoNo
Pro ($16.99/mo)YesYes (19 languages)Yes (host-initiated)By request
BusinessYesYes (19 languages)Yes (host-initiated, admin controls)By request
EnterpriseYesYes (19 languages)Yes (full admin controls + audit)Yes

Cloud transcript languages (19)

English, Spanish, French, German, Italian, Simplified Chinese, Russian, Ukrainian, Japanese, Korean, Vietnamese, Dutch, Portuguese, Arabic, Polish, Romanian, Swedish, Turkish, Danish. If your meeting is in any other language (Hindi, Thai, Indonesian, Hebrew, etc.), Zoom’s cloud transcript won’t produce useful output.

AI Companion: a different product

AI Companion is bundled with Pro+ tiers but is a separate feature from the cloud transcript. It produces a structured summary(key points, action items, next steps) rather than a verbatim transcript. The host must explicitly start it during the meeting — it doesn’t run automatically. Distribution is host-controlled: if the host doesn’t share, participants get nothing.

If you’re the host

Path A: Native Zoom transcript (Pro+)

  1. In Zoom web settings → Recording → enable Cloud recording
  2. Toggle Audio Transcript on
  3. Start your meeting and click Record → Record to the Cloud
  4. End the meeting normally — processing starts when the session closes
  5. Wait 5-30 minutes (depending on length), then go to web portal → Recordings → click the meeting → download the .vtt file

Path B: Local recording on any tier

Free hosts (or Pro+ hosts who want to skip cloud) can record locally. The recording saves as .mp4 (video + audio) and .m4a (audio-only) in your Zoom folder (usually ~/Documents/Zoom/ on Mac, C:\Users\[you]\Documents\Zoom\ on Windows).

No native transcription happens for local recordings. Upload the .mp4 (or .m4a if you don’t need video) to any transcription service. DeluxeScribe handles both formats with speaker labels and 99 languages.

Path C: AI Companion summary (Pro+, complementary)

During the meeting, click AI Companion in the toolbar to start summary generation. After the meeting ends, the summary appears in your Zoom account and (if you opted in) gets emailed to participants. AI Companion runs on a different language set (30+ supported) than the cloud transcript (19) — so for languages between the two lists, you can get a summary but not a verbatim transcript natively.

If you’re NOT the host

This is the case the “how to transcribe a Zoom meeting” listicles ignore. You attended a meeting, you’d like a transcript, but you don’t control the recording. Three options, in order of effort:

1. Ask the host to share

Usually the right first step. If the host had cloud recording on, they can forward the recording link or share the .vtt transcript file. If they had Audio Transcript disabled, they can’t produce one after the fact from their cloud recording, but they can download the MP4 and share it with you to upload yourself.

2. Record your side locally during the call

Plan ahead — this requires you to start recording before or during the meeting. Check your state’s recording consent laws first.In one-party-consent states (most US states), you recording yourself in a meeting you’re in is usually fine. In two-party-consent states (California, Florida, Pennsylvania, others), you may need to notify all participants.

  • Mac: QuickTime → File → New Screen Recording → Options → choose audio source (Mac system audio requires BlackHole or Loopback as a virtual audio router, not built-in)
  • Windows: Game Bar (Win+G) records the active window with system audio, or OBS Studio for more control
  • iPhone: Control Center → Screen Recording → press-and-hold to enable microphone. Note this records your mic, not the meeting audio directly

3. Personal note-taking bot

Fireflies, Otter, Granola, and similar tools can be added to your calendar and will auto-join Zoom meetings as a participant, recording the meeting on your behalf. The host will see a bot in the participant list — some hosts find this rude or refuse to let unknown participants join. Use with judgment and disclose.

Once you have a local recording from any of the above, upload it to a transcription service. DeluxeScribe accepts the MP4, M4A, or M4V output from any of these recording methods.

Native Zoom vs DeluxeScribe — honest comparison

Zoom native cloud transcriptDeluxeScribe
CostIncluded with $16.99+/mo Zoom plan60 min free, $10/mo for 1,200 min
Languages1999
Speaker labelsYes (session-based; degrades on cross-talk)Yes (audio-based diarization)
Export formatsVTT onlyTXT, DOCX, PDF, SRT, VTT, JSON
Editor for fixing errorsNo (raw VTT only)Yes (in-browser, plus speaker renaming)
RetentionTied to cloud recording (admin policy)Your account, your retention
SetupOne-time toggle per Zoom accountUpload file when needed
Works when you’re not the hostNoYes (any audio/video file you have)
HIPAA / BAAAvailable on Zoom enterprise; request requiredNot HIPAA-compliant

When Zoom native wins

  • You’re the host
  • You’re on Pro tier or higher
  • Your meeting is in one of the 19 supported languages
  • You only need the raw VTT
  • You’re fine with your org’s retention policy

When DeluxeScribe wins

  • You’re on Zoom Free
  • You’re not the host
  • Your meeting is in a language Zoom doesn’t support natively
  • You need DOCX, PDF, or SRT output
  • You need to fix errors in an editor before sharing
  • You want an independent archive that outlives the cloud recording

Upload a Zoom recording and read it in minutes

60 minutes free, no credit card. Drop your Zoom MP4 or M4A and get a searchable transcript with speaker labels, timestamps, and 6 export formats.

Retention — your transcript dies with the recording

Zoom cloud recordings are kept per admin policy, not on a Zoom-wide default. Many organizations set retention to 30, 60, or 120 days, after which both the recording and the attached transcript are permanently deleted. There is no “keep the transcript, delete the recording” option — they’re bound together.

If you need to reference the transcript past your org’s retention window:

  • Download the .vtt file as soon as it’s available and store it in your own system (Drive, Notion, a folder)
  • Or download the MP4 and re-transcribe later through a third-party service that keeps your transcripts as long as your account is active
  • For research where data must be auditable years later, don’t rely on the org’s Zoom retention — own the archive

Privacy and BAA

Not legal advice. Consult your compliance team for specifics.

HIPAA

Zoom offers a Business Associate Agreement (BAA) for HIPAA-eligible accounts, but it’s not active by default. Your Zoom administrator must request and execute it. Without a BAA, Zoom can’t process Protected Health Information lawfully even if the call audio is technically secured. AI Companion processing has its own data-sharing controls that admins must configure.

DeluxeScribe is not HIPAA-compliant. Do not upload audio containing PHI. For clinical use, either activate a Zoom BAA and use Zoom native, or self-host Whisper.

Attorney-client privilege

Many firms restrict cloud transcription for privileged content. Check your firm’s data-handling policy before uploading either to Zoom (if your firm doesn’t have a BAA-equivalent confidentiality agreement) or to a third party. Self-hosted Whisper is often the only option that satisfies strict policies.

GDPR

If your Zoom meeting includes EU residents, both the recording and the transcript are personal data. You need a lawful basis for processing. Zoom publishes a DPA and offers EU-region processing in some configurations. Third parties (including us) act as processors under your control; ask for a DPA before uploading EU-resident audio.

Common gotchas

  • You’re on Pro+ but no cloud recording option. Your Zoom admin has disabled cloud recording at the org level. Contact your admin or use local recording.
  • Transcript shows “Speaker 1, Speaker 2” instead of names. Zoom labels by session participant; if speakers join from a phone or join late, they may be labelled generically. Use a third-party service that diarizes from audio for cleaner labels.
  • VTT won’t open in Word. VTT is plain text; open it in any text editor, or upload to a transcription service for conversion to DOCX/PDF.
  • Meeting in an unsupported language → empty transcript.Zoom’s native transcript only handles 19 languages. For everything else, export the MP4 and upload to a multi-language service.
  • You’re on Free wondering where the transcript is.Free tier doesn’t include cloud recording or transcription. Local recording on Free produces an MP4 you can transcribe with a third-party tool.
  • The meeting ended abnormally and there’s no recording.If Zoom didn’t process the recording (network drop, app crash), there’s no cloud transcript either. Local backup recording (Zoom’s “Record on this computer” option alongside cloud) is the insurance policy.
  • AI Companion didn’t run.Host has to start it manually during the meeting — it doesn’t auto-enable even on Pro+. If nobody started it, no summary exists post-hoc.

How this page was verified

Zoom feature claims were verified against Zoom Support KB0064927 (audio transcript languages), KB0058013 (AI Companion), and the Zoom pricing page as of June 2026. Retention is set per Zoom account by the administrator — there is no published default; we note that honestly rather than fabricating a number. BAA availability confirmed via Zoom Trust Center. Comparison tool pricing was captured June 2026 from each vendor’s public pricing page. We don’t cite blanket “99% accuracy” figures common in competitor copy because they aren’t sourced to a published benchmark.

Frequently Asked Questions

Does Zoom transcribe meetings automatically?

On Pro tier and above, yes — Zoom's cloud-recording transcript is automatic if you enable Audio Transcript in Settings. You get a VTT file alongside the recording, usually within 30 minutes of the meeting ending. Free / Basic tier doesn't include cloud recording or transcription. AI Companion summaries are a separate Pro+ feature that the host has to explicitly start during the meeting.

Is Zoom transcription free?

No. Cloud-recording transcription requires Pro ($16.99/mo) or higher. Free-tier Zoom doesn't include cloud recording, so there's nothing for the native transcript to attach to. Local recording on Free tier produces an MP4 you can transcribe with a third-party tool — DeluxeScribe handles MP4 with 60 minutes free.

How do I get a transcript from a Zoom recording I didn't host?

If you weren't the host, you don't get access to Zoom's native transcript — that goes to the host. Three options: (1) ask the host to share the recording or transcript, (2) record your own copy locally during the call (Mac: QuickTime + system audio; Windows: OBS or Game Bar), (3) deploy a personal note-taking bot like Fireflies, Otter, or Granola that joins meetings as a participant. With a local recording you can upload to any transcription service.

What languages does Zoom's transcript support?

Zoom's cloud-recording transcript supports 19 languages as of 2026: English, Spanish, French, German, Italian, Simplified Chinese, Russian, Ukrainian, Japanese, Korean, Vietnamese, Dutch, Portuguese, Arabic, Polish, Romanian, Swedish, Turkish, and Danish. AI Companion summaries cover 30+ languages. If your meeting is in a language not on Zoom's list, export the MP4 and upload to a multi-language service.

How long does Zoom keep recordings and transcripts?

Retention is set by your Zoom admin, not by Zoom directly. Many organizations keep cloud recordings 30-120 days, after which the recording AND the attached transcript are permanently deleted. If you need a long-term archive, download the .vtt file as soon as it's available and store it yourself.

Is Zoom transcription HIPAA-compliant?

It can be, but only with an active Business Associate Agreement (BAA) on the Zoom account, which is not enabled by default. Contact your Zoom account administrator or Zoom Trust if you need a BAA. AI Companion processing also has its own data-sharing toggles that admins can configure. For attorney-client privileged content, check your firm's policy on cloud transcription before either path. DeluxeScribe is not HIPAA-compliant — don't upload PHI to us.

Why is my Zoom transcript showing wrong speaker labels?

Zoom assigns speaker labels using session participant identity, which works well for distinct turn-taking but degrades during cross-talk, when someone uses multiple devices, or when guests join as 'Phone User'. The transcript labels may also lag the actual speaker change by a few seconds on busy meetings. For research or legal use where speaker attribution matters, a third-party service with diarization-from-audio is more reliable than Zoom's session-based attribution.

What's the difference between Zoom's AI Companion summary and the cloud transcript?

Different products. The cloud transcript is a verbatim VTT file of who said what, attached to the recording, available on Pro+. AI Companion produces a structured summary (key points, action items, next steps) — also Pro+ but the host must start it explicitly during the meeting. Both can run on the same call. The transcript is for reading; the summary is for skimming.