How To Produce Quality Media Localization
If you’re working with Wordbee Translator, check out our new online subtitle translation feature for previewing video subtitle translations. Wordbee also offers end-to-end solutions for some of the world’s largest media localization teams.
According to Nimdzi, “Media localization is not only glamorous, it’s arguably the most attractive segment to be in the language services market right now.” Supporting this statement with data, media localization vendor ZOO Digital has seen an increase in revenue of 143% within two years. Additionally, four out of five fastest-growing LSPs in 2018 are media localization providers.
Judging from these numbers, media localization is a clearly growing industry. But rest assured, more companies will tap into it — on both the technology, as well as the translation vendor’s side.
So how do you keep up with the competition, you ask? Quality. Media. Localization.
What Does It Take To Produce Quality Media Localization?
When looking at a more general idea of quality media localization, it can be described as the transition of audio and video content into other languages and for other cultures and countries. The process can include dubbing, subtitling, access, and other media services. Media localization is quite a complex process. Lorena Santonocito, who is a freelance audiovisual translator for both SDI Media and ZOO Digital, shared with us a couple of skills that are fundamental to deliver a good localization project:
- Effective communication
- A deep knowledge of the market and audience you’re talking to (type of clients, cultural and linguistic elements, geographical aspects, values, etc.)
- “Balanced Creativity,” meaning the capacity to convey the exact thought and adapt it to a specific audience without losing the original message
Lajka Birkedal Hollesen, co-owner and partner at Localised Web and subtitler for Verboo and Pixelogic, explains that audio-visual translation and subtitling differs from traditional text localization in the sense that it’s more restricted and context dependent. To create quality media localization you need to take into account audience specific aspects, such as:
- Different reading speeds for adults and children
- The variation of positioning and formats of subtitles for specific languages
- Adherence to the client’s specifications regarding character limitation, subtitle duration, cueing, and on-screen text
Same as Lorena, Lajka also mentioned that a good understanding of the target audience is crucial to having quality media localization. Other key factors are skilled native subtitlers specialized in media localization, scalable tools that support many file and video formats, as well as integration of assisted tools such as KNP (Key Names and Phrases) databases. Same as with all types of localization, technology plays an important role in producing quality work. However, the type of technology can vary. In media localization Neural Machine Translation (NMT), for example, is not yet suitable even though it works well in many other industries; the reason being that the content that needs to be localized is oftentimes dialogue, which contain humor, sarcasm, dialect, and more, said ZOO Digital CEO Stuart Green at SlatorCon London this year. So what technology supports translators most with this type of work?
What Translation Technology Is Most Important For Media Localization?
Using CAT tools with translation memory and translation glossaries to keep consistency in projects is very useful. However, sometimes the work of a media localization professional can be quite creative so they don’t always make use of these tools, admits Lorena. Most important to her are quality assurance tools that detect errors/mistakes, inconsistent segments, or content that has not been translated.
Emphasizing this point, Lajka states it’s more challenging to optimize the process of producing high quality media localization than the translation of mere text-related content assisted by state-of-the-art CAT tools. A number of factors need to be taken into account in media localization and a pre-translation feature of recognized terms or sentences in a subtitle would not necessarily be useful as it could result in reading speed violations. The current equivalent to a traditional CAT tool’s translation memory in today’s media localization is the above-mentioned KNP tool, which can be used throughout the project. The higher uniqueness of the language-specific media content and the differences in condensation and dialogue strings also make it less beneficial to use translation memories.
In order to meet clients’ needs a variety of interchangeable tools that support different video and file formats must often be used in media localization as it’s difficult to find one system that meets all requirements. Sometimes even non-localization-native tools are being used, such as Adobe Premiere, a video editing software, which supports many video formats and has an intuitive interface for template creation.
Other tools that are specifically being used for subtitling include WinCaps, Swift, EZTitles, Aegisub, and Titlebee. However, the list for both free and paid software is quite extensive, and some clients may even provide specific software to work with.