Speaker identification makes it easier to follow conversations and know who said what in your meeting transcripts. Here’s how it works in Voicenotes:
Regular notes vs. meetings
If you’re recording a regular note, speaker identification will not be available.
To get speaker identification, make sure the meeting toggle is turned on before you start recording.
This tells Voicenotes to treat the recording as a meeting, so speakers can be separated in your transcript.
How speaker identification works
With meetings: Voicenotes uses AI models to separate voices into different speakers (e.g. Speaker 1, Speaker 2). This works well, but it’s an estimate and not always perfect.
With meeting bots: For the most accurate results, invite the Voicenotes bot to join your Zoom, Google Meet, or Microsoft Teams call. This way, speaker names are pulled directly from the meeting, so you’ll see actual participant names instead of generic labels.
Recording with the meeting bot (recommended)
Joining meetings with the Voicenotes bot is currently available on the web app only. We’re working on bringing this to mobile soon.
Here’s how to record with the bot:
Step 1. Open the web app
Go to voicenotes.com/app and log in.
⚠️ This feature is currently available only on the Voicenotes web app.
Step 2. Select 'Record Meeting'
Click the blue button to start recording.
Step 3. Select “Join meeting as a bot”
From the menu that appears, choose Join meeting as a bot.
Step 4. Paste your meeting link
Insert the Zoom, Google Meet, or Microsoft Teams link and click Join.
The bot will join your meeting and your transcript will include accurate speaker names.
Summary
Regular notes → no speaker identification.
Meetings (with toggle on) → AI separates speakers as Speaker 1, Speaker 2.
Meetings with bot (web app only) → participants’ actual names appear in your transcript (most accurate).
Still have questions? Reach out to us at [email protected] and we’ll be happy to help.