There is a significant difference in cost and complexity
With video and audio taking on an increasingly greater role in eLearning, translation decisions regarding how to format and deliver audio content become a critical piece of the localization puzzle. Those decisions can determine how quickly your translations get completed and the value of the finished product to your learners. In this piece we’ll look at those issues and others to help you make the choice that works best for your needs.
The decision to choose subtitling or voice-over recording and editing for translated multimedia content is a tough one. Besides the often considerable difference in cost and turnaround time, there are issues of cultural acceptance and how effective one option is over the other in clearly delivering information. On top of those important issues, everything becomes even more complex when there are multiple languages involved and/or less common languages. Sometimes the availability of native-speaking voice talent can be a major challenge and even a blocker that forces the issue.
eLearning translation workflows for subtitling and voice-over production
Typically, both paths start with a transcription of the audio content, which then goes through the normal translation and review processes. Each line of content relative to a particular scene is tagged with an ID so the translation can be matched to its required location in the video. This is the first place where issues can arise.
Character counts vary widely from language to language
Translated content often has a different character count, sometimes shorter, sometimes longer. German, for example, can be up to 30% longer because their equivalent words are often much longer. When you have limited time and space on the video this can be a big problem that may require revising the original content for length. You are often limited to 35 characters per line and two lines per scene. Otherwise it will be difficult to read and may block important visual content. By the way, my previous sentence is 78 characters so it represents the upper limit of what you can display.
In addition to length, people speak much faster than most can read so the transcribed text may simply be too long to work. The answer to this is to write the script with the understanding that it will be read as text as well as audio. An experienced video writer can make this adaptation and it will make the final product much easier to consume in your target languages.
Specialized subtitling vendors may be required
Once the length has been dealt with, the titles must be created in video editing software and there must be access to the source video files to create a new version with the subtitles. It is not unusual for this to be handled by a third party vendor who specializes in subtitling. This can be important because the choices of text size, font and color can depend on the background image and adjusted in order to be readable.
The transcriptions that start the process are also typically done by a third party transcription process and may be supplied by the client or the language service provider (LSP) can manage that process.
Foreign language voice talent and production are a whole next level
For voiceovers, the transcription process is the same but the goal of the process is to create an audio track in the target language(s). The script has timecodes amended to each line or scene dialog so the translated and recorded audio files can be inserted into the correct scene. The above-mentioned issues with length can also arise here.
Once the script is translated, there is a selection process for the voice talent. Make of female? Casual or businesslike? Generally the tone used in the original should drive this process. The challenge here is that it can be quite difficult to find professional talent in lesser used languages. Generally speaking it is not advisable to use company executives or employees from global distribution partners to record soundtracks. Experienced talent will be much faster and their pacing will be steadier, making it much easier to edit their recordings into the completed video. There are online portfolio sites where you can listen to sample recordings of available talent. Your LSP may be able to point you to these
The recording and editing of the translated content is generally handled by a specialist firm in the target country and managed by the project manager at the LSP. Again, finding these resources can be difficult in smaller or emerging markets.
The fundamental issues in choosing which route to take are resources and time
When you get down to it, the choice is often about money and timelines, followed by resource availability. In cases where many languages are involved there will be significant savings in time and money by going with subtitles. You also limit the need for hard to find talent, recording, and editing of audio tracks. The tradeoff is that the user experience may suffer somewhat. It is important to note that these voiceover steps require time and will affect your delivery timeline.
It’s usually not the LSP making this decision, but we may make a recommendation
Given that one option is considerably more costly than the other, when we are quoting a multi-language video translation where voiceover is requested, we may quote both options so the client has an understanding of the difference. We do not have a preference but having the backstop of the subtitle option does help when you’re faced with a high quote for voiceover. It eliminates the cost of the talent, the recording, and the audio editing which, multiplied by the number of languages can skyrocket. It really does come down to money and time. But there is another consideration: cultural preferences.
In some countries subtitles are common and accepted while in others it can be insulting to not provide an audio track in their language. This is a localization issue that may require some research before making the choice between your two information delivery options in video translation.