You talk with your client.The notes write themselves.
Record the conversation on your phone or laptop. Within a few minutes you have a written transcript, the agreements listed and a draft email to your client. The recording never leaves your own computer or server.
It ran at our place first.
We record client conversations with our own app on the laptop or the phone, connected to our AI systems. The AI button of common recording apps and recorders sends the recording to the provider's servers, and that is not an option for businesses in healthcare or for a client under confidentiality. So at our place the transcription runs fully locally, with Whisper large-v3 on our own computer, and an agent then turns it into notes with action points. The recording never leaves the machine.
Does this sound familiar?
- 01After an hour-long conversation the write-up costs another half hour, so you push it back.
- 02Agreements made verbally mean something different to both sides a week later.
- 03The AI in ordinary recording apps uploads everything to a server you do not know.
- 04In healthcare, notary work and advisory work, a client conversation is not allowed to leave the building.
What you get.
- A recording app for your phone and laptop, connected to your own AI system, no new device needed
- Fully local transcription (Whisper large-v3), in Dutch, with your clients' names as hints
- Notes with decisions, open questions and action points per person
- A draft email with the agreements, ready to send
- Storage per conversation in a fixed folder, searchable
Operating it yourself says more than a description.
Start the recording of a fictional intake conversation. You see the conversation transcribed line by line, followed by the notes, the action points and the draft email. This is a demo on sample data, not client work.
Demo · prepared in advance · illustrative · no live model
No recording yet. Start the recording to see the conversation transcribed live.
Illustrative demonstration with examples prepared in advance. No live AI call and no API key needed; the answers are fixed and contain no invented figures or names.
Four steps, in this order.
We sit in
We look at which conversations you have, who takes part and what needs to happen with the notes afterwards.
We build alongside
The recording app and the processing run on your own computer or server, separate from your other systems. If you stop, nothing is stored anywhere else.
You keep the last step
Whatever the system produces is first ready as a draft. Only when you give the go-ahead does anything leave the building.
We connect only once it works
Once it proves right in practice, and only then, we connect it to your systems for good.
Lead time: one to three weeks for a working first version.
Whisper large-v3 · whisper.cpp · Own recording app · Python · Local or shielded AI
What people ask us about this.
Q1Does my recording go to the cloud?
No. The transcription runs on your own computer or server. For writing the notes we choose together: a model that also runs locally, or a shielded business AI environment with agreements on data security. The recording itself never leaves your machine.
Q2Do I need a special recorder?
No. You record with your phone or laptop through an app we build for it. If you already have a recorder, that works too, as long as its cloud feature is switched off.
Q3Am I allowed to record a conversation?
Always tell the other person beforehand that you are recording and why. If you work with health data or under professional secrecy, local processing is precisely the way to stay within the GDPR, because the recording does not end up with an external processor.