Log your thoughts and ideas with quick voice notes. Easily share ideas and brainstorm in a voice document with AI assistance for progress tracking and more.
This is the 2nd launch from VoiceTaking. View more
VoiceTaking v2: The meeting note taker
Launching today
Transcribes your meeting, without bots joining your call
Every other meeting note taker sends a bot into your call. VoiceTaking doesn't. You're already in the meeting, so it captures it right from your browser — meeting audio and your mic on separate channels, transcribed live and multilingual, so the notes always know who said what. Pick a template and the summary takes that shape, with action items, owners and due dates. Meet, Zoom, Teams, Discord. Desktop app coming soon!
Hey Product Hunt 👋
VoiceTaking started as a voice notes app. You pressed record, spoke, and got a transcript you could shape with AI. People used it for brainstorms, journaling, dictating drafts on a walk.
Then I looked at what they were actually recording. Over and over: meetings. Someone propping a laptop next to a speakerphone. Recording a Google Meet call on their phone and uploading the file afterwards. One user told me she typed through every client call anyway, because the transcript only ever caught her half of the conversation — hers.
That last one bothered me for weeks.
The fix everyone reaches for is a bot. Connect your calendar, send a robot into the call, let it sit in the participant list. It works, but the cost is quietly enormous: someone has to explain it, someone has to let it past the waiting room, and people genuinely speak differently once a visible recorder joins a candidate interview. It also wants calendar access and an account on every platform you use.
Which felt backwards. You're already in the meeting. Your computer is already playing the audio.
So VoiceTaking now captures the call from the browser you're already in. You share the tab, your mic covers your side, and both get transcribed live on separate channels — so the notes always know who was talking. Nothing joins your call. No integration, because there's nothing to integrate.
The build changed shape twice. I started on Whisper for transcription and it was fine for single-speaker recordings, but it couldn't keep two live channels straight. Moved the whole pipeline to Deepgram nova-3, which handles multichannel and diarization natively and code-switches mid-sentence — my team drifts between languages constantly and I wanted the transcript to follow.
The other thing I only learned by using it: muting yourself inside Zoom is completely invisible to the browser. I sat through a call assuming I was muted while it happily transcribed me. There's now a Mute mic control that actually works.
Free to try, no card, nothing to install.
I'd really like to hear where it breaks for you — especially if you run meetings in a language or a setup I haven't tested. Reply here, I'm around all day.
Report
No reviews yetBe the first to leave a review for VoiceTaking
LaterOn.email