We use real people, whose first language is English.

We only transcribe what has been said. We never edit or sanitise the transcript and we include in the text instances of hesitation, repetition, emotion and silence, to give a 3-D impression of the speaker. Robust Anglo-Saxon doesn’t bother us.

Each transcriber’s work is read back, by a different operator, against the original recording to ensure accuracy.

Automatic voice to text software can, at first blush, appear effective, both in terms of time and cost. Using our system of working, however, can mean the difference between exoneration or a custodial sentence.

We are mindful that English is a melange of many languages, with other cultures and influences adding to the mix. English relies heavily on context. For example: the sound “sight” could also be “cite” or “site” depending on context. This contextual concept, coupled with idiomatic use of English, many non-native speakers don’t always appreciate and AI often disregards.

We concentrate on accuracy, presentation and confidentiality.

As we’re real humans we look forward to discussing your specific requirements.