What every sector shares
Whatever is being recorded, it is almost always about people: clients, candidates, council members, students, sources. That is why Murmel works the same way for everyone. Audio and transcripts are processed and stored in the Netherlands, you choose whether a recording stays or is deleted automatically after 1, 7, 30 or 90 days or a year, and recordings are not used to train the model unless you opt in. The model is trained on Dutch speech, with English as a second language, and makes the fewest errors of the tested models on open Dutch benchmarks (the measurement). You teach it your terminology and names with a custom vocabulary.
For organisations with several users there are team accounts with a shared pool of minutes and one invoice. You arrange a data processing agreement via contact. What an hour of audio costs is on the pricing page.
Each sector in brief
- [Law firms](/transcriberen-voor-advocaten): client meetings and hearings, with care for professional secrecy.
- [Education](/transcriberen-voor-onderwijs): search lectures and subtitle course material, English mixed in too.
- [Council meetings](/raadsvergaderingen-transcriberen): long meetings with many speakers as the basis for minutes.
- [HR](/transcriberen-voor-hr): job and exit interviews with consent and a short retention period.
- [Consultants](/transcriberen-voor-consultants): client interviews and workshops quickly searchable.
- [Journalism](/transcriberen-voor-journalisten), [research](/transcriberen-voor-onderzoek) and [government](/transcriberen-voor-overheid): check quotes, justify in a data management plan, work digitally sovereign.
Weighing Murmel against another service or a typing agency? Compare puts prices, languages and data storage side by side. All features are listed under features.