How to Record a Webinar on a Mac (and Get a Transcript)

Webinars have a specific cruelty: they’re the meetings where you most need notes and have the least power to make them. A vendor demo you’ll be quizzed on, a compliance training with dates that matter, a paid workshop moving faster than you can type — and the record button belongs to someone else. The host might share a replay. It might arrive in a week, behind a login, with an expiry date. Or never. If you’ve tried to record a webinar on a Mac as an attendee, you’ve hit the wall from both sides: the platform won’t let you, and the notetaker bots that work for normal meetings get stopped at registration.

The way through isn’t fighting the platform — it’s ignoring it. Every webinar, whatever it runs on, ends as audio playing out of your Mac. macOS can capture that audio at the system level, transcribe it live, and summarize it, all locally, without the host’s permission model ever being involved. This post covers why the usual routes fail, the local workflow step by step, and what to do when a replay file does show up.

In this guide7 sections

Why the normal routes fail for webinars

The platform route. Zoom Webinars, Microsoft Teams webinars, and GoTo Webinar all reserve recording for the host or organizer. As an attendee you get no record button, and post-event transcripts or replays exist only if the host enables them, only in the form the host chooses, and only when the host gets around to sending them. Even when a replay does arrive, it’s a video file — not a searchable transcript, not action items.

The bot route. Cloud notetakers — Otter, Fireflies, and the rest — mostly work by joining the call as an extra participant (their bot-free desktop apps send what they capture to the vendor’s cloud). Webinars are structurally hostile to that: there’s a registration wall with your name on it, often a waiting room, and a hard split between panelists and attendees. Many hosts block bots outright; on some platforms an attendee-side bot can’t hear anything worth transcribing anyway. The bot approach assumes the meeting is a room anyone can walk into, and a webinar is precisely not that. It’s the same structural dead end that closes the bot route for Slack huddles and FaceTime.

The screen-recording route. QuickTime can’t capture system audio without a third-party virtual audio driver, and a full screen recording of a one-hour webinar produces a multi-gigabyte video you’ll never rewatch. You didn’t want a video; you wanted the words.

What all three routes miss is that the audio is already on your machine. You don’t need to be let in — you’re in.

How to record a webinar on a Mac, step by step

The clean way to do this is capturing the webinar app’s own audio output via ScreenCaptureKit, the macOS framework built for exactly this. That’s what Dictanta does — audio only, scoped to the one app playing the webinar, with live transcription running as it plays. The workflow:

1. Join the webinar normally. Zoom app, Teams, or a browser tab — it doesn’t matter. There’s nothing to configure inside the webinar platform, because the capture never touches it.
2. Start a recording in Dictanta (⇧⌘R, or the optional global shortcut ⌃⌥⌘R if you’ve enabled it in Settings). The source picker offers three options: My voice, This meeting, and Another app.
3. Pick the source. If the webinar runs in Zoom, Teams, Webex, Slack, or a browser (Safari, Chrome, Arc, Brave, Firefox, and Edge are all detected — which covers Livestorm, Demio, YouTube Live, and every other browser-based platform), This meeting finds it automatically. For a desktop app Dictanta doesn’t auto-detect, such as the GoTo app, bring it to the front and choose Another app, which captures the frontmost app’s audio.
4. Grant the permission once. The first capture asks for macOS’s Screen & System Audio Recording permission. Despite the name, Dictanta records audio only — never video, never a pixel of the screen. The mechanics are covered in how to record system audio on a Mac.
5. Watch the transcript stream. Transcription runs live while the webinar plays, on-device via Apple’s SpeechAnalyzer, so by the final slide the full transcript already exists. It autosaves every 30 seconds — if your Mac sleeps or the app crashes an hour in, the recording is recovered and marked “Recovered” instead of lost.

One detail to know: both system-audio options mix your microphone in — that’s by design, because in a normal meeting your own voice is half the record. In a webinar where you never speak, it simply means your mic’s silence is part of the mix. If you ask a question during Q&A, that’s captured too, which is usually exactly what you want.

Which source to pick, by platform

The two options behave identically once recording starts; the only difference is how the source is identified. This meeting recognizes the call automatically, while Another app simply takes whatever app is frontmost when you start — so for a GoTo session, click into the GoTo window, then start the recording. Either way the capture is locked to that one app’s audio: a Slack ping or a YouTube video in another window won’t bleed into the recording the way it would with a mic pointed at your speakers.

From transcript to something you’d actually use

When the webinar ends and you stop the recording, Dictanta generates a summary on-device with Apple Foundation Models: a TL;DR, action items with owners and due dates, decisions, and open questions. For a webinar, that maps to “what was announced, what you’re supposed to do about it, and what to follow up on.”

Summary lines that show a timecode are audio-anchored — hover one and playback jumps to the exact moment it came from; lines Dictanta can’t source show no timecode instead of a guessed one. For a dense one-hour session, this is the difference between notes you trust and notes you re-verify: the speaker quoted a price or a deadline, the line claims a number, one hover confirms it in their own voice. The recording auto-renames from the summary topic, so an hour later it’s filed as the subject that was actually discussed, not “Recording at 14:02”.

From there, everything exports: Markdown for your notes app, PDF or Word for the colleague who asked what they missed, SRT if you’re subtitling clips, JSON if you’re piping it somewhere. Search covers titles, transcripts, and summaries in the app, with titles and transcripts in Spotlight too, so “what was that Q4 pricing change” is findable months later. And because every step ran locally, a webinar recorded on a plane Wi-Fi connection transcribes just as well as one recorded at a desk — the whole pipeline works offline once the language model is on your Mac.

If the host does send a replay

Sometimes the replay file shows up after all. That’s not wasted — it’s an import. Drag the file into Dictanta on the Mac (or press ⌘O), and a batch transcription pass produces the same transcript, summary, and anchored bullets as a live recording. Supported formats include M4A, MP3, WAV, and AIFF, and imported files transcribe the same way any audio file does.

Two things make imports worth knowing about even if you always record live. First, imports don’t consume the free quota — Dictanta’s free tier counts your first 3 completed recordings, and imported files aren’t recordings, so you can transcribe a backlog of replay files at no cost. Second, it covers the webinars you genuinely couldn’t attend: if a colleague forwards the replay, you get the searchable transcript without having been there.

A note on permission

Recording locally means the platform doesn’t know and doesn’t notify anyone — which makes the judgment yours. For most webinars the stakes are low: it’s a broadcast, you’re taking notes on something presented to you. But check the event’s terms for paid workshops and courses, where the content itself is the product — recording for your own notes and redistributing the material are very different acts, and the second one is a copyright problem regardless of the tool. For anything interactive where you speak, the same disclosure courtesy applies as in any recorded meeting.

Requirements: Dictanta needs macOS 26.4 on Apple silicon (M1 or later), with Apple Intelligence turned on.

When to use this

Use local capture whenever you’re an attendee and the notes matter: vendor demos, trainings, investor calls run as webinars, paid workshops (within their terms), conference sessions streamed to your browser. It’s the only approach that works the same across every platform, doesn’t depend on the host’s settings, and produces a transcript the moment the event ends rather than whenever the replay email arrives. If you’re the host, your platform’s native recording may serve fine — though the same on-device workflow gives you a transcript without a bot in the participant list and without uploading your event to a transcription vendor.

One adjacent case: the same event attended in person. System-audio capture is Mac-only, but a conference talk in a physical room is a mic recording — one tap on iPhone or iPad, live captions as the speaker talks, same on-device summary at the end. That workflow, including Apple Pencil markup on the transcript, is covered in lecture transcription on iPad. The Mac workflow in this post is specifically for events that reach you as audio played by an app.

Your first 3 recordings are free with every feature unlocked and no length cap — the next webinar on your calendar is a complete test.

FAQ

Can you record a webinar as an attendee?

Not through the webinar platform — Zoom, Teams, and GoTo all reserve recording for the host or organizer. What you can do on a Mac is record the audio the webinar app or browser tab is playing on your own machine. That capture happens locally, works on any platform, and doesn’t depend on the host sharing a replay.

Do notetaker bots work in webinars?

Usually not. Bot-based tools like Otter or Fireflies attend as an extra participant, and webinars are built to control exactly who gets in — registration walls, waiting rooms, and panelist-versus-attendee roles often block a bot or make it useless. Capturing the audio on your own Mac sidesteps the join problem entirely because nothing has to enter the event.

What if the host sends a replay file afterward?

Import it. Dictanta accepts M4A, MP3, WAV, AIFF, and more via drag-and-drop or ⌘O, runs a batch transcription pass on-device, and generates the same summary with audio-anchored bullets. Imports don’t count against the 3-recording free quota, so transcribing replay files is free even before you subscribe.