Transcription seems easy. You take a recording of a speech, sit someone at a computer, they listen to the recording and type it up. Voilà!
Except it’s never that easy. One hour of speech takes on average 6 hours to transcribe. But many speeches at international conferences are made by non-native speakers. You may have a French doctor speaking in English at a medical conference. His accent when speaking English makes it hard to understand what he’s saying. Even harder if he uses acronyms.
So one hour of speech can easily become 8 or 10 hours of transcription work.
Then there are the ad-libs. A speaker may have their speech written up beautifully. But when on the podium, they make jokes, get lost in the text, mumble, thank the organisers – none of this is planned, usually it’s said unclearly, often it’s not relevant to the content of the speech. But the transcriber has to transcribe it anyway, and filter out the ‘Umms’ and the ‘Ahs’.
And of course, we’re assuming that the microphone was pointing in the right direction, was switched on, that there’s no background noise, and that the recording started at the beginning.
There are a lot of variables which can make transcription into a very difficult task.
To make sure that we provide you with the best possible transcription service, all our transcriptions go through a team of three professionals: a transcriber, an editor, and a proofreader. All of them, naturally, are native speakers of the language in which the speech was given.
We have built up a database of professional transcribers since our first St Petersburg International Economic Forum in 2009. Every applicant is required to sit a complicated test, which will help us understand if they have the skills we require, and whether they will make a better transcriber, editor, or proofreader.