24/7 meeting transcription on a private server

In-house systemClient: Flowara AI (in-house system)Sector: Meetings and documentation

An in-house system that turns meeting recordings into notes with a summary and open tasks, separating who is speaking, with no monthly fee to third parties.

The problem

After a meeting with a client there are hours of audio that nobody listens to again. Transcription services charge a monthly fee and keep the audio outside your control.

What was built

  • Input from a phone, a computer or a shared folder: the recording is uploaded and the system does the rest.
  • Transcription in Spanish and English, separating who is speaking at each moment.
  • A final note with a summary and open tasks, filed by client and by month.
  • An index per client and a notice when it finishes.
  • One job processed at a time, with stable memory use even when a meeting lasts hours.

What you can see

A synthetic demo with a sample audio file and the note it produces. Still to be prepared.

Shown with synthetic data, never with real client data.

What it shows

Turning conversations into documentation you can search, on a private server and with no monthly fee per use.

Other cases