Methods and workflow
Cloud transcription
Transcription where the audio is uploaded and processed on the vendor's servers.
Cloud transcription means the audio is sent to a vendor that processes it on their servers and returns the text.
This is the most common form, and for most ordinary recordings it is the right one.
The advantages
- Fast regardless of your machine. The processing happens on hardware built for it.
- No installation. Works in the browser.
- Always the newest model. The vendor updates without you doing anything.
- Scales. Twenty hours of audio does not take twenty times as long as one.
The four questions you need answered
Before you upload anything concerning other people:
- Where is the audio stored, and for how long? Inside the EEA, or not?
- Is it used to train the vendor’s models? Many free tiers are financed exactly that way.
- Is there a data processing agreement? Without one your organisation cannot lawfully use the service on personal data.
- Who has access at the vendor?
For meeting summaries from ordinary work meetings this is manageable with an agreement in place. For health data, client conversations and personnel cases, local transcription is what removes the question entirely.