Several years ago we got a call from a large multinational oil company looking for some foreign language voice-overs. They had a 15 minute video and were interested in recording an off-camera narration in several languages. I was speaking directly to the producer at their in-house recording studio.
So far so good. I became concerned when their estimated time to record this 15 minute video was exactly 15 minutes. When I asked about the translations, I was told they did not have translations and did not need them. When I asked how they were going to record German, Russian, French and Japanese without a written script for the talent to read, in 15 minutes no less, I got a lecture. “That’s what the talent does. They supply a simultaneous translation. They hear the English and then speak their simultaneous translation. I thought you’re the translation company. Don’t you know this is how you use simultaneous translation to do voice-over recordings?”
I made sure it was not April fools day.
I had never before or ever since heard of anyone attempting to listen to an English audio track and spontaneously, in perfect sync with the video, provide a flawlessly spoken simultaneous translation.
Anyone who has ever listened to simultaneous interpreters knows that they are by the very nature of their work “running behind” the speaker they are interpreting and trying hard just to keep up with them, because they have to know what s/he is saying before they can start to relay the message. While it is called simultaneous interpreting or simultaneous translation, it is actually really more like “almost simultaneous” or “time delayed, catch up with the speaker interpreting.” There are frequent false starts, back tracking and corrections, the kind of pauses that come with looking for the right word or waiting for the speaker to get to the object or verb at the end of the sentence, not to mention the invariable “…umms” and “…ehhs” and then the rushing to try to keep up with the speaker. It is far from the kind of nice, clean and well-paced read you want for your corporate promotional video.
And with respect to recording time, even the very best VO talent will need many multiples of the run time to get a flawless read. For lock-to-picture recording with a listen back for QA purposes we estimate that 5–7 minutes of finished material will take experienced VO talent one hour of recording time.
This particular client, though, would not be dissuaded. When I pointed out that any retakes would invariably result in a different read, and that there would be no control over the content, there was incredulity. How could one English text possibly yield more than one interpretation? If that was the case, then clearly the narrator clearly not suitable for the job of providing a simultaneous translation.
It is worth noting that many narrators are not translators or even interpreters. They are primarily actors and not necessarily linguists. And while simultaneous translators are widely considered the most skilled linguists in the profession, they do not consider themselves voice-actors.
Politely, I declined to get involved with their project.