I’ve recently been working on some translations of courses created by another developer. He had already translated one of the courses into Chinese. I had to do translations of two courses into Turkish, Russian and Polish, none of which I know.
The courses were in Articulate Storyline. For each, I was provided with the Storyline file in English, a script translated into the target language, and one or two audio files for each course containing the voiceover for all the slides of that version. All the courses contained videos for which I had to provide subtitles in the target language. The three translations differed widely in their approach to audio and script.
From this experience I discovered some things that make it harder and some that make it easier, which I’ll offer here in case you have a similar job but don’t have a process in place already.
1 – the audio files
Think it’s hard finding your way around a foreign town? Try finding your way around an audio file in a language you know nothing about! You have a 30m waveform in front of you with little idea where one slide ends and the next begins.
The quality of the audio files varied. For one of the courses, I was given three audio files which covered the entire course. One of them was recorded in stereo for no particular reason – there was only one voice – so I resaved it as mono to reduce file size. One of the three had been recorded under different conditions – perhaps a different room, or the narrator sitting further from the mic, from the other two. I had to do quite a bit of tweaking to make them sound consistent. This narrator would hesitate, retake sentences or even parts of sentences, and even occasionally comment to someone else in the room, so there was a lot of editing to remove these hesitations. I highlighted the section of the audio specific to the slide I was working on, and saved the selection as a separate file to import into Storyline.
At least this one helpfully spoke the slide numbers in English, which enabled me to find where I was in the course. One of the others said them in the target language, which was no use to me, and the other just didn’t say them! For Russian I had to go to the lengths of learning the Russian alphabet (via Memrize) just so I could figure out where what I was reading matched what I was hearing
2. the Storyline slides
I had a couple of problems with the slides. One was that Storyline authors often uses layers and states to show and reveal text on a slide. If the translator is working from a Storyline Word output, they may not see these and not give a translation. The second is inevitable – some translations end up in much longer text than the English version. So your screen design may have to change to accommodate them. You may have to use a different font from the original – one of my courses used a ‘whiteboard marker’ font which didn’t have Russian or Polish characters. I had to find a similar font and supply it to the original author.
One translator forgot to include all the prompts – Next, Back, Submit, Menu etc. Another didn’t supply the slide titles, so the menu appeared in English.
4. Checking the translation
In one translation the translator typed the translation below the slide. Often it wasn’t clear which translated term corresponded to which English term. This is where Google translate proved handy. I’d put a group of Russian or Polish words or sentences into the left-hand window of Google translate; the English translation was often wacky but close enough to give me an idea what was meant. By hitting Return in the left-hand window to isolate words or phrases, the English translation got paragraph breaks in the corresponding places. Gradually by trial and error I could figure out what phrases corresponded to the phrase I was looking to replace. It’s OK to use Google translate to get your orientation like this, but you’d never want to use it for text the translator hasn’t provided. That way lies madness.
This is not to criticise the translators – everyone was learning how to do this as we went. With all this in mind, here are some guidelines that might be useful to you if you’re working with translations of elearning courses.
Supply the translators with
* an up to date script that corresponds to the version of the course they’re looking at
* clear indication what needs to be audio and what is text only (for example video subtitles script is text only, no need for audio)
* slide titles
* all on-screen text that needs to be translated including
* buttons – Back, Next, Submit
* any content hidden in layers or mouseovers
* question feedback
Translators should supply
* One audio file per scene/section, not one for the whole course
* The file should be mono, preferably a high quality mp3, not WAV, to reduce file size
* Slide numbers spoken in English before each slide’s content
* Gaps of ten seconds between slides
* Retakes, hesitations, coughs, page turnings and repetitions edited out before the file is handed over
* Where slides are complex with many small text boxes, a table with the English terms and equivalents is easiest to work with
* Text should as far as possible be the same length as the English, especially with labels appearing on a slide – to replace an eight-letter English word with a 20-letter word often means something else on the screen may have to be removed, or resulting translated text may be too small to read