AI notetaker for lawyers and legal teams
Wispr Notetaker is an AI notetaker for lawyers that labels speakers in the final transcript, spells case names and legal terms from your personal dictionary, and lets you set how long transcripts are kept. Legal teams get a record they can quote back weeks later, attributed to the right person.
What a legal team actually needs from a meeting record
In legal work, people go back and read the record again. Saving time is not why it matters. A client intake call, a witness prep session, a partner-level matter review: the notes from those conversations get read back weeks later, quoted in an email, or used to reconstruct what a client asked for. If the transcript spells a party's name three different ways, or attributes a concession to the wrong person, everything downstream inherits the error.
So there are really only three questions worth asking. Does the transcript name the speakers, or does it say "Me" and "Them"? Does it know how to spell the names on the matter? Who decides how long the transcript lives, and can it be deleted for good? Wispr Notetaker is built around those three questions, so here is how each one behaves in practice.
One thing to be plain about up front: an accurate record is not a substitute for telling people you are recording. That part is on you, and no tool changes it.
A day in the life of a legal team's meetings
The meeting mix in a legal team is unusually varied, and each type wants a different output.
- Client intake: names, dates, and the client's own phrasing, captured verbatim enough to quote
- Matter review with a partner: decisions and who owns the next step
- Opposing counsel calls: what was proposed, what was refused, and by whom
- Internal case strategy: a working record nobody outside the team should see
- Vendor and e-discovery calls: acronyms, tool names, and deadlines
- In-person conversations in the office, on a shared mic
Wispr Notetaker records on your device instead of joining as a bot, so it works the same way whichever app the call is in. Named speaker labels do vary by platform, covered below. Nothing appears in the participant list, but people on the call should still know they are being recorded. Google Meet, Zoom, Microsoft Teams, Webex and Slack huddles are covered, and in-person meetings work on a shared mic. Back-to-back meetings hand off cleanly, which matters on a day of stacked half-hour calls. A call that was never on the calendar gets detected, with a one-tap start.
Meeting type, output, and where the notes end up
Sharing happens one note at a time, to one person at a time. Access comes by link or email invite to named people. There are no shared folders and no project spaces, which is a real constraint if you were hoping to mirror your matter structure inside the tool.
Your typed notes and the generated summary stay separate. Editing or deleting one leaves the other alone, and your own edits are never silently overwritten. For a lawyer who annotates during a call, that separation is the difference between a working document and a mess.
How the transcript gets names and terminology right
Two mechanisms do most of the work, and neither of them involves reading your screen.
Context from the invite and your apps. Wispr Notetaker goes into the meeting knowing who is on the calendar invite, and it draws on your calendar, Gmail, and Slack. That is why an uncommon name spelled correctly on the invite comes back spelled correctly in the transcript.
Your personal dictionary. This is the same dictionary Wispr Flow uses for dictation, and Wispr Notetaker inherits it. Terms you already taught Flow show up in your meeting transcripts from day one. Add matter names, party names, statutes you cite constantly, and the acronyms your practice group lives on; entries take effect immediately and sync across your devices.
During the meeting you get a live transcript without speaker names. Once the meeting ends, Wispr Notetaker goes back over the recording and the refined transcript carries named speakers. Anyone it could not identify, you name once afterward and it applies across the whole transcript.
Speaker attribution, and where it is strongest
Attribution is the part most worth testing before you trust it on a sensitive matter, because it varies by platform.
- Google Meet in Chrome, Edge, Brave, or Chromium: named speaker labels appear automatically, and the label self-corrects shortly after a handoff
- Zoom in the native app: named attribution is supported, though Zoom's active-speaker detection is less reliable than Google Meet and Microsoft Teams. Browser-based Zoom is not supported
- Microsoft Teams in the native app: named attribution is supported
- In person on a shared mic: Speaker 1 and Speaker 2 until the refined transcript resolves them
Wispr Notetaker also uses your Mac's system-default mic, and a mid-meeting device change switches over almost immediately. If the other participants' audio drops, the session continues on your mic alone and picks the channel back up, so a network wobble does not cost you the meeting.
Retention, deletion, and the trust questions a legal team will ask
This is where a legal team's diligence usually starts, so here is the honest version.
Wispr Notetaker encrypts meeting audio and keeps it only temporarily, on your device and sometimes in cloud storage. It uses that audio to build your transcript, let you resume a meeting, and check quality. After that limited period it removes the audio automatically. Wispr Notetaker does not create voiceprints or biometric profiles of you or anyone on the call.
Transcript retention is a separate setting and it is yours. Keep transcripts indefinitely, which is the default, or have them auto-delete after one, seven, 30, 90, 180, or 365 days, set in Wispr Notetaker settings with notes sync on. If your firm's admin sets the policy centrally, the setting shows as managed by the admin.
On compliance: Wispr Flow is SOC 2 Type II and ISO 27001 compliant on every plan, not gated behind an enterprise tier, and reports are available to anyone through the Wispr Flow Trust Center. HIPAA-ready covers dictation with a signed BAA, and Wispr Notetaker is not available on a BAA account. Model training is a setting called "Improve the model for everyone," shown during onboarding and changeable any time in your data controls; it is on by default on Flow Free and Pro, and off by default on Enterprise.
Keeping your existing meeting history when you switch
Nobody in a legal team wants to strand a year of matter notes in an old tool. You do not have to.
Coming from Granola or Otter, one-click import brings your history into Wispr Notetaker. Coming from anything else, the route is to keep both: connect your old notetaker and Wispr Notetaker to Claude, and one chat searches the old archive and your new notes together. Name both tools in the prompt when you want the older material, because AI tools tend to default to a single source.
What import does not do is rebuild your old tool's sharing setup, because Wispr Notetaker shares per note to named people rather than through folders or workspaces. Nothing needs deleting on the other side either. Run both notetakers on one real matter review, compare the final transcripts, and see whether the names and the attribution hold up.
Who this fits, and who should wait
A good fit if you work on a Mac, your meetings are mostly in English, and the thing you care about is an accurate attributed record you control. It fits solo practitioners and small teams especially well, because SOC 2 Type II and ISO 27001 are on every plan and speaker identification is available on Flow Free.
Wait if any of these describe you.
- Your work requires a signed BAA, since Wispr Notetaker is not available on a BAA account
- Your team is on Windows, which is coming soon
How to test it on one real matter
Pick a meeting where accuracy actually matters, and check the four things that decide it.
- Before the call, add the party names, the matter name, and your practice group's acronyms to your personal dictionary
- Tell everyone on the call that you are recording, then start Wispr Notetaker from the notification
- During the meeting, type your own notes alongside the transcript, and ask the in-meeting chat what you missed if you step away
- Afterward, read the refined transcript first: check the spellings, check who said what, and name anyone who came back as "Speaker N"
For a long deposition-style session, set the recording length before you start: the default is two hours, and you can choose 30 minutes, one, two, or three hours.
One meeting tells you most of what you need to know. If it does not hold up, you have learned that in an hour rather than a quarter. Try Wispr Notetaker on your next client call and read the refined transcript before you decide.
Frequently asked questions
Can Wispr Notetaker record an in-person client meeting?
Yes, on a shared mic. Everyone shows as Speaker 1 and Speaker 2 during the meeting, and the refined transcript resolves them afterward. You can also name anyone it could not identify once, and the name applies across the whole transcript.
How do I make sure case names and legal terms are spelled correctly?
Add them to your personal dictionary in Wispr Flow, which Wispr Notetaker inherits. Entries take effect immediately and sync across your devices. Add both singular and plural forms of a term, and use misspelling correction for words that come back wrong the same way every time. Starred entries get higher priority.
Is Wispr Notetaker available for Windows and on Enterprise plans?
Wispr Notetaker runs on Mac today, with Windows coming soon and mobile on the roadmap. It is available on Flow Free and Pro, and coming soon to Enterprise. If your firm needs SSO, SCIM, audit logs, and org-wide retention enforcement now, those are Enterprise features and Wispr Notetaker is not part of that plan yet.
Will I lose my meeting history if I switch from Otter or Granola?
Your history comes with you: one-click import from Granola or Otter
Is Wispr Notetaker HIPAA compliant?
HIPAA-ready covers Wispr Flow dictation with a signed BAA, which an individual can accept in Settings. Wispr Notetaker is not available on a BAA account, so if your practice requires one, Wispr Notetaker is not the right fit yet. Separately, Wispr Flow is SOC 2 Type II and ISO 27001 compliant on every plan.
Do meeting participants know I'm recording?
Nothing joins the call, so recording happens on your device. Wispr Notetaker keeps a "get consent when transcribing others" notice in front of you the whole time it records, and links out to guidance. Your organization can require you to confirm consent before recording starts, and you can add a Notetaker notice to your calendar invites.
Who can see a transcript I have not shared?
Nobody. A note stays unshared until you send it. Access comes two ways: a share link, which shows the summary only, or an email invite to a named person, which is what gives access to the full transcript. Invited people get viewer access and can see the owner and the other invitees.
