The average knowledge worker spends 31 hours per month in meetings. A significant chunk of that time is spent taking notes, or worse, reconstructing what was said after the meeting ends. Meeting transcription apps promise to solve this by automatically capturing and organizing what everyone says. But the landscape is messy. Some tools join your Zoom calls as a bot. Others record system audio locally. Some generate structured notes with action items. Others dump a wall of text and call it a transcript.
We tested the most popular meeting transcription tools that work on Mac in 2026: Steno, Otter, Jamie, Fireflies, and Zoom's built-in transcription. This guide breaks down what each does well, what it does poorly, and which one fits your workflow.
Two Approaches to Meeting Transcription
Before comparing individual tools, you need to understand the two fundamental approaches to meeting transcription on Mac. They are different enough that choosing the wrong category wastes more time than choosing the wrong app within a category.
Bot-based transcription
Tools like Otter and Fireflies join your video calls as a participant. They show up in the meeting with a name like "Otter.ai Notetaker" or "Fireflies.ai." They record audio directly from the call platform, identify speakers using the platform's participant list, and produce transcripts with speaker labels, timestamps, and summaries.
The advantage: fully automatic. Connect your calendar, and the bot joins every meeting without you doing anything. Speaker identification is reliable because it matches voices to named participants. The disadvantage: everyone in the meeting sees the bot. Some people find this intrusive. Some companies ban third-party bots from calls. And if you are in a sensitive meeting, having a third-party AI bot recording can create trust and compliance issues.
Local audio capture
Tools like Steno capture audio directly on your Mac. They record what comes through your speakers (the other participants) and your microphone (you). No bot joins the call. No one else knows you are transcribing unless you tell them. The audio stays on your device or is processed locally.
The advantage: invisible and private. Works with any audio source, not just video calls. You can transcribe a podcast, a lecture, or an in-person conversation the same way you transcribe a Zoom call. The disadvantage: you have to start and stop recording manually. Speaker identification relies on audio analysis rather than participant names, so it is less precise.
Quick Comparison
| App | Approach | Price | Speaker ID | Summaries | Calendar Sync | Offline |
|---|---|---|---|---|---|---|
| Steno | Local audio | Free / $4.99/mo | Yes (audio) | Structured notes | No | Yes |
| Otter | Bot | Free / $16.99/mo | Yes (platform) | Yes + action items | Yes | No |
| Jamie | Local audio | Free / $24/mo | Yes (audio) | Yes + executive summary | Yes | No |
| Fireflies | Bot | Free / $18/mo | Yes (platform) | Yes + CRM sync | Yes | No |
| Zoom built-in | Platform | Free with Zoom | Yes (platform) | Basic | N/A | No |
Detailed Reviews
Steno Meeting Mode
Steno is primarily a dictation app. You speak, it types. But it also has a meeting mode that captures system audio and microphone input simultaneously. Start it before or during a call, and it records everything being said on both sides. When the meeting ends, you get a transcript with speaker separation and the option to generate structured notes.
What Steno does well in meetings is the same thing it does well for dictation: accuracy on specialized vocabulary. If your meetings involve medical terminology, legal language, financial jargon, or technical terms, Steno handles those better than generic transcription tools because it adapts to your profession. A product meeting where someone says "we need to deprecate the v2 API and migrate to gRPC" gets transcribed correctly. A medical team huddle where someone mentions "metformin titration" lands right.
Steno also processes audio on your device. The recording never leaves your Mac. For teams that handle sensitive information, whether that is patient data, legal strategy, financial plans, or proprietary product discussions, this matters. You do not have to worry about a third-party AI company storing your meeting recordings on their servers.
What Steno does not do: It does not join Zoom or Google Meet as a bot. It does not sync with your calendar and auto-join meetings. It does not label speakers by name pulled from the meeting platform. It does not integrate with CRMs or project management tools. If you need a fully automated, hands-off meeting transcription pipeline, Steno is not the right tool.
Best for: People who want meeting transcription that is private, accurate on specialized vocabulary, and works alongside their existing dictation workflow. Especially good for one-on-one calls, interviews, and small team meetings where you do not need automatic calendar integration.
Otter
Otter is the most popular meeting transcription tool for a reason: it works well and it is easy to set up. Connect your Google or Outlook calendar, tell it which meetings to join, and a bot named "Otter.ai Notetaker" shows up in your Zoom, Google Meet, or Teams calls. After the meeting, you get a transcript with speaker labels, highlights, action items, and a summary.
The free tier gives you 300 transcription minutes per month with basic features. The Pro plan at $16.99/month adds unlimited minutes, custom vocabulary, advanced search, and the ability to export transcripts. The Business plan adds admin controls, analytics, and priority support.
Otter's speaker identification is strong because it pulls participant names directly from the video platform. When John Smith speaks, the transcript says "John Smith" not "Speaker 2." Over time, Otter learns voice profiles and can identify recurring participants even in audio-only recordings.
Best for: Teams that want fully automated meeting transcription with reliable speaker identification. Companies that use Zoom or Google Meet for most meetings. People who want searchable meeting archives.
Honest limitation: The bot joining your meeting is visible to everyone. Some clients and external partners find this uncomfortable. The free tier is too limited for heavy use. All audio is sent to Otter's cloud servers for processing. And Otter only works for scheduled video calls. It cannot transcribe an ad-hoc in-person conversation or a podcast you are listening to.
Jamie
Jamie takes the local audio approach but adds cloud-powered AI summaries. It captures audio from your Mac without joining the call as a bot. After the meeting, it sends the audio to its servers for processing and returns an executive summary, action items, and a full transcript. No one in the meeting knows Jamie is running.
Jamie's summaries are genuinely good. They read like something a competent executive assistant would write: a paragraph summarizing what was discussed, a bulleted list of decisions made, and a list of action items with who is responsible. You can customize the summary format to match your company's meeting notes template.
At $24/month for the Pro plan, Jamie is one of the more expensive options. The free tier gives you 5 meetings per month, which is enough to test but not enough to rely on.
Best for: Managers and executives who need polished meeting summaries they can share with stakeholders. People who do not want a bot in their meetings but still want AI-generated notes.
Honest limitation: Expensive. Audio is sent to Jamie's cloud servers for processing, so it is not truly private. Speaker identification from local audio is less reliable than bot-based tools. Limited integrations compared to Otter and Fireflies.
Fireflies
Fireflies is similar to Otter in approach: a bot joins your video calls, records, and transcribes. Where Fireflies stands out is in integrations. It connects to Salesforce, HubSpot, Slack, Notion, Asana, and dozens of other tools. After a meeting, it can automatically push action items to your project management tool, update CRM records, and post a summary to a Slack channel.
Fireflies also offers a "Topic Tracker" feature that lets you define topics and automatically highlights when they come up in meetings. If you want to know every time a competitor is mentioned across all your team's meetings, Fireflies can surface that.
Best for: Sales teams that want meeting transcripts synced to their CRM. Organizations that need cross-meeting analytics and topic tracking. Teams that use Notion or Asana for project management and want action items pushed automatically.
Honest limitation: The bot participant issue is the same as Otter. The free tier is very limited. The sheer number of integrations and features can feel overwhelming if you just want a simple transcript. Audio is cloud-processed.
Zoom Built-in Transcription
If you use Zoom for most of your meetings, you already have meeting transcription. Zoom's built-in feature generates live captions during the call and a full transcript after. The transcript includes speaker labels (for signed-in participants), timestamps, and a searchable text record. Zoom AI Companion, available on paid plans, adds meeting summaries and action items.
The quality is decent for everyday business English. It struggles with accents, cross-talk, and specialized terminology more than dedicated tools. But it is free, it requires no setup, and it does not add a bot to your meeting.
Best for: Zoom-only teams that want basic transcription without installing anything new. Organizations that already pay for Zoom and want to use what they have.
Honest limitation: Only works in Zoom. If you also use Google Meet, Teams, or have in-person meetings, you need a separate solution for those. Transcription quality is below Otter and Steno. Summaries from AI Companion are basic compared to Jamie or Fireflies. No integrations with external tools.
Choosing the Right Tool for Your Workflow
The right meeting transcription tool depends on three things: how many meetings you have, how sensitive the content is, and how much automation you need.
If you have 2-4 meetings per day and want zero effort
Use Otter or Fireflies. Connect your calendar, set it to auto-join, and forget about it. Every meeting gets transcribed, summarized, and archived without you lifting a finger. The bot visibility is the price you pay for automation. Most people get used to it and so do their colleagues.
If your meetings involve sensitive information
Use Steno. Audio stays on your device. No cloud processing of your recordings. No bot in the meeting that signals to everyone that the conversation is being captured by a third-party AI. This matters for healthcare teams discussing patients, legal teams discussing strategy, and executive teams discussing confidential plans.
If you need polished summaries to share
Use Jamie. Its executive summaries are the best in class. You can customize the format, and the output reads like professional meeting minutes rather than an AI transcript. Worth the higher price if meeting notes are a significant part of your job.
If you only use Zoom
Start with Zoom's built-in transcription. It is free and already there. If the quality is not good enough, then consider Otter or Steno. Do not pay for a third-party tool until you have confirmed that Zoom's free option does not meet your needs.
A Practical Workflow for Meeting Notes with Steno
Since Steno takes a different approach from the bot-based tools, here is what using it for meetings actually looks like:
- Before the meeting: Open Steno on your Mac. Switch to meeting mode. Select your system audio as the input source so it captures what other participants say.
- Start the meeting: Join your Zoom, Google Meet, or Teams call as usual. Click record in Steno. Nobody in the meeting sees anything different.
- During the meeting: Steno captures audio in real time. You can see a live transcript if you want, or minimize it and focus on the conversation.
- After the meeting: Stop recording. Steno generates the full transcript with speaker separation. You can then use Steno's AI actions to generate structured notes, extract action items, or summarize the discussion.
- Share or file: Copy the notes into your team's wiki, Notion, Slack channel, or email. Since Steno is also a dictation app, you can dictate follow-up emails and messages right from the same tool.
The whole process adds maybe 30 seconds to your meeting routine: one click to start, one click to stop. The tradeoff compared to Otter or Fireflies is that you have to remember to start and stop. You do not get calendar integration or auto-join. For some people, that manual step is a dealbreaker. For others, the privacy and control are worth it.
Meeting Transcription Accuracy: What to Expect
No meeting transcription tool is perfect. Here is what affects accuracy and what you can realistically expect in 2026:
- Audio quality is the biggest factor. A good external microphone in a quiet room produces dramatically better transcripts than a laptop mic in a noisy coffee shop. If you care about transcript quality, invest in a decent microphone before you invest in a better transcription app.
- Cross-talk kills accuracy. When two people talk at the same time, every transcription tool struggles. The best tools handle it gracefully by noting that speech overlapped. The worst tools produce garbled nonsense. Encourage turn-taking in meetings you plan to transcribe.
- Accents and non-native speakers. All tools handle standard American and British English well. Performance varies with other accents. If your team is international, test tools specifically with your team's voices before committing.
- Specialized vocabulary. General-purpose tools like Zoom's built-in transcription stumble on medical, legal, and technical terms. Steno's profession-aware vocabulary handles these better. Otter allows custom vocabulary on paid plans.
- Realistic accuracy range: 85-95% in good conditions. That means 5-15 errors per 100 words. For a one-hour meeting with roughly 9,000 words spoken, expect 450 to 1,350 words to need correction. That sounds like a lot, but most errors are minor (a "the" becomes "a") and do not change meaning. Plan to spend 5-10 minutes reviewing any transcript before sharing it.
Privacy and Compliance Considerations
Meeting transcription raises real privacy and legal questions that you should think about before turning on recording.
Consent laws: Many jurisdictions require all-party consent for recording conversations. In the US, this varies by state. California, for example, requires all parties to consent. New York only requires one party (you). If your meetings include participants from different jurisdictions, the strictest law applies. Bot-based tools make consent easier because everyone can see the bot. Local recording tools put the responsibility on you to inform participants.
Data residency: Cloud-based tools like Otter, Jamie, and Fireflies store your meeting recordings and transcripts on their servers. If your organization has data residency requirements (common in healthcare, finance, and government), check where each tool stores data. Steno processes audio on your device, which avoids this issue entirely.
HIPAA and protected information: If your meetings involve patient health information, you need a tool that is either HIPAA-compliant or keeps data entirely on your device. Most cloud-based meeting transcription tools are not HIPAA-compliant on their standard plans. Steno's on-device processing sidesteps this. For more on privacy-first transcription, see our voice-to-text guide.
When Steno Is the Right Choice vs When It Is Not
We make Steno, so we want to be honest about where it fits and where it does not.
Choose Steno for meetings when:
- Privacy matters. Audio stays on your Mac.
- Your meetings involve specialized vocabulary that generic tools get wrong.
- You also use Steno for dictation and want one tool for both.
- You do not want a bot joining your calls.
- You have a small number of important meetings rather than back-to-back calls all day.
Choose something else when:
- You have 6+ meetings per day and need fully automatic transcription. Use Otter or Fireflies.
- You need meeting transcripts synced to your CRM. Use Fireflies.
- You need executive-ready summaries to share with stakeholders. Use Jamie.
- You only use Zoom and want free. Use Zoom built-in.
- You need cross-meeting search and analytics across your entire team. Use Otter Business or Fireflies Business.
The Bigger Picture: Meeting Notes as Part of Your Workflow
Meeting transcription is not an end in itself. The point is to capture decisions, action items, and context so you can act on them later. The best tool is the one that fits into how you already work.
If you live in Notion, pick the tool that pushes notes to Notion. If you live in Slack, pick the tool that posts summaries to Slack. If you work solo and just need a record of what was said, pick the simplest tool that gives you accurate text.
Steno fits into a broader workflow because it is not just a meeting tool. It is a dictation app that also handles meetings. After your call ends, you can use the same app to dictate follow-up emails, write up your own notes, or draft messages to your team. One tool for all the words you need to produce, whether you are in a meeting or at your desk writing. For a full overview of dictation tools on Mac, see our best dictation app for Mac comparison.
Frequently Asked Questions
Can Steno transcribe Zoom meetings?
Yes. Steno captures system audio on your Mac, so it transcribes audio from Zoom, Google Meet, Teams, or any other video call. It does not join the meeting as a bot. You start Steno's meeting mode before or during the call, and it records what plays through your speakers and microphone.
What is the best free meeting transcription app for Mac?
Zoom's built-in transcription is free for all Zoom users and works well for Zoom-only meetings. Otter offers a limited free tier with 300 minutes per month. There is no truly free, full-featured meeting transcription tool for Mac. Steno offers a free trial if you want to test meeting mode alongside dictation.
Do meeting transcription apps record my calls without permission?
Bot-based tools like Otter and Fireflies join calls as visible participants, so everyone sees them. Local tools like Steno capture audio on your device without joining the call. In both cases, you should inform participants that transcription is active. Many jurisdictions legally require consent from all parties.
Can meeting transcription apps identify different speakers?
Most dedicated meeting tools identify speakers. Bot-based tools like Otter and Fireflies use the platform's participant list for accurate speaker names. Local tools like Steno separate speakers by voice characteristics but may label them as Speaker 1 and Speaker 2 rather than by name. Accuracy depends on audio clarity and how distinct the voices are.
Is it better to use a meeting bot or local transcription?
Meeting bots are better for automation, calendar integration, and speaker identification. Local transcription is better for privacy, invisible recording, and flexibility across audio sources. If you have back-to-back meetings all day and want zero-effort transcription, use a bot. If you have sensitive meetings and value privacy, use local transcription.
How accurate is meeting transcription in 2026?
The best tools achieve 90-95% accuracy in clear audio with native English speakers. Accuracy drops with background noise, heavy accents, and cross-talk. No tool is perfect. Expect to spend a few minutes reviewing any transcript you plan to share. Good audio equipment makes a bigger difference than the choice of transcription tool.
Meeting transcription tools save time, but they do not replace paying attention. Use them to capture what was said so you can focus on what it means.
Try Steno for meeting transcription and everyday dictation: trystenofast.today. One tool for meetings, emails, documents, and everything else you need to write.