Teams, Zoom, Meet and in person
The sound is recorded on your computer. Which conferencing program is running makes no difference.
100% local · GDPR-compliant without the cloud
AI meeting minutes, without your meeting leaving the room
SpeechToWork records your meeting and then writes the transcript and the AI meeting minutes: summary, decisions, action items with owners, open questions. Everything is created on your computer.
- ≈ 9 sanalysis per minute of meeting in our test.
- 2 tracksyour microphone and the sound from Teams, Zoom or Meet.
- 0 botsjoin your meeting. Recording and minutes are created on your PC.
An example
What you say. What appears.
You sayA meeting about the website relaunch, three participants, 58 minutes.
In your programDecisions
- Go live on 15 November
- Copy by 31 October from Ms Berger
- Photos from Mr Shaw
Action items
☐ Me: send the revised quote (tomorrow)
Open questions
- Do we need an online shop?
Why AI minutes from the cloud are a risk
Most AI note takers join your meeting as a bot or upload the recording to their servers. That puts the spoken words of your clients, applicants and colleagues with a service provider, often in the US. For HR conversations, client meetings or price negotiations, that is hard to justify.
SpeechToWork needs no bot. On your PC it records two tracks: your microphone and the sound coming from Teams, Zoom or Meet. The analysis happens locally. That makes it an AI note taker for Teams calls and for in-person meetings alike.
What you get after the meeting
- A transcript with timestamps and speakers. Your voice is marked as “Me”, the others are told apart.
- Minutes with four sections: summary, decisions, action items, open questions.
- Action items as a checklist with owner and deadline.
- Both as a text file in one folder per meeting, ready to share or file.
In our test, one minute of meeting was analysed in about nine seconds. An hour-long meeting is therefore minuted in a few minutes.
Record only with consent
Before you record, everyone taking part should know and agree, whatever tool you use. In many countries, including Germany, recording a private conversation without consent is a criminal offence, and data protection law expects people to be informed. SpeechToWork reminds you before every recording. The advantage of local processing: you never have to explain which provider ends up with the recording, because none does.
Your advantage
Why SpeechToWork fits this job.
Action items, not walls of text
The minutes separate decisions from action items and state who does what by when.
Feeds into your activity log
In the evening, SpeechToWork combines dictations and meetings into a report of your day.
Privacy
Local means local. No compromises.
Speech recognition and the language model are installed on your PC and run there. What you dictate ends up in your program and nowhere else. How the data flow works.
No upload
Audio and text stay on your computer. There is no server listening in and no AI provider in the background.
GDPR made simple
No third party processes your dictations. So there is no data processing agreement to sign and no international transfer to assess.
Works without internet
Once installed, SpeechToWork works offline. Only the licence check needs a connection.
Good to know
Frequently asked questions
Can the AI write minutes from a Teams meeting?
Yes. SpeechToWork records the sound that reaches your PC from Teams, plus your microphone. You need no bot in the meeting and no approval from your IT team in Microsoft 365. The same works with Zoom, Google Meet and Webex, and with meetings around a table in your office.
Do I need a headset?
It is recommended, but not required. With a headset, your voice and the voices of the others are cleanly separated. Without one, your microphone also picks up the speakers. For that there is a switch that mutes your own track while other people are talking, so nothing is transcribed twice.
How good is the speaker separation?
Your own voice is reliably recognised through your microphone. A model separates the other participants by their voices and numbers them as Speaker 2, 3 and so on. With similar voices or poor sound quality, mix-ups can happen. The transcript itself remains complete, so nothing that was said goes missing.
How long can a meeting be?
There is no fixed limit. The recording is written to your hard drive as it runs, so a long meeting cannot fill up the memory. The language model condenses long transcripts section by section and then combines them into one set of minutes with decisions, action items and open questions.
Coming soon
Coming soon. Then try it for 14 days.
SpeechToWork is about to launch. As soon as the first version is ready, you can download it here: one click, one file, no form.
Coming soonFor Windows 10/11 (64-bit). The download will be available here as soon as it is ready.
- All features, no payment details
- Ends automatically, nothing to cancel
- Your dictations never leave your computer, not even in the trial
Download
One installer for Windows, straight from our server.
Install
A double click is all it takes. No administrator rights needed.
Choose “Try for 14 days”
On first start: enter your name and business email, no payment method.
On first start, SpeechToWork downloads the language models once (4 to 6 GB). Already have a licence key? Enter it on first start. What is transferred in the process is explained in our privacy policy.