Audio Translation Guidelines from STAR Translation
Choosing the Script’s Tone (and the actor)
One of the first decisions in audio translation is to decide how the voice translation tone should be.
Should it upbeat and youthful, professional and accurate, even paced or high energy? Our recommended mode for most translations is to use a professional newscaster-style, which conveys solid and accurate information of your products.
The second decision is whether to use male or female voices. This is ultimately the customer’s decision. Where possible, we choose female voices for product marketing material. You should also choose your actor carefully as you may need to update scripts in the future — you need to ensure their availability when required — voice talent is always booked in advance.
For technical marketing, there are a number of ways to pronounce your products terminology.
Actors will by default for IT products use a US-style pronunciation for IT terminology such as iPod, Internet, NAS, DHCP etc, as this is the way words are commonly pronounced around the world.
Occasionally some clients will have specific pronunciation-style requests for new technology or key product names. We will always strive to determine if such requirements are required in advance.
Developing your script – Timing
When you record your original English voice script you should remember the following points for translation.
- Translation typically will be longer, sometimes up to 30%. Therefore you should leave additional time spacing in your audio or video links to incorporate this potential change.
Translation language order can also change. For some languages the order of words in sentences can change. So for example if you have the text “to see a demo, click here”, you cannot rely on the timing of the term “click here” in your video.
Default rule: record segments of your script and leave appropriate extra silences for the translators.
Audio and Video Timing Adjustments – Tracking
When syncing a video clip to an audio clip, we recommend you record as above segments of text rather than have one single continuous recording.
The English version is actually broken into a string of key audio segments marked with silence markers. These silences can then be expanded by the translation team to ensure the translation matches your video.
If you have four sequence shots in your video, it would be best to have four audio segments joined as a single audio file. This ensures the translation process is smoother and delivers higher quality results.
Working with Source Files – Timing
For all audio translations we only work with source files from you. This ensures that our timing sequences match yours exactly. Playing files from CDs or streaming video can impact the timing based on CD speed or the speed of a weblink.
The Audio Translation Process
The process we utilize for translation of audio is as follows.
- First we obtain a written copy of your script.
- The script is then translated into the target language.
- Any terminology issues or questions are identified at this stage. We also review the text with respect to general timing and length of text. If the text is too long, we will adjust the translation accordingly with advice from the you.
- We select the voice artist for your audio translation.
- We review the video or audio for key timing points and translation suitability.
- We translate/record each of your files in segments matching timing and tracking information as appropriate.
- When completed, we check the audio track against your original for timing and translation quality before finally delivering the completed translation back to you.
The STAR Team