Insatiable demand, talent shortages and the English master script
With demand for localization services growing larger and increasingly diverse, more and more workflows are starting with a non-English master script, writes Take 1 CEO, Louise Tapia.
As streaming has evolved during the past few years, we’ve witnessed a phenomenal increase in the demand and range of localization services. Increasingly, content platforms recognize that localization is integral to optimizing the value of their programming – and, consequently, their investment in these services is also growing.
But as welcome as this undoubtedly is for Take 1 and other companies in the field, it does raise some critical questions about how localization could – and should – be delivered going forward. One of these concerns is the tradition of using an English master script as the starting point from which all other language versions are created – in effect, becoming a blueprint for all localization.
Now think back a decade when the market was dominated by shows originating in English, and you could make a strong case for this standard approach to master scripts. But to the benefit of audiences everywhere, streaming has opened up a new, much more internationalized era of viewing. Customers are now just as likely to be binge-watching shows that were originally produced in French or Korean – as the smash-hit status of, respectively, “Call My Agent” and “The Squid Game” confirms.
With demand for new programming soaring during lockdown, the audience for foreign language content has continued to grow. But with delivery schedules becoming more intense and localization resources under pressure even before the Covid-era content boom, it’s time to consider whether the current approach to localization remains fit for purpose.
Source language > target language?
Logically enough, one of the suggested changes is that more content be translated directly from the original language to the target language. For example, rather than translating a Korean film into English and then into French you’d translate directly from Korean into French. Going directly from source language to target language makes it easier to align linguistic nuances, among other benefits. But dig a little deeper and it soon becomes clear that this approach does bring its own challenges.
First and foremost, there are already significant localization talent shortages. Because English is so ubiquitous around the world, it’s easier to find talent if English is always included in the language pairings, for example: Korean-English and English-French rather than Korean-French. Given the current supply-demand imbalance, we’re unlikely to establish source-target language pairings that satisfy every language requirement any time soon.
There is also a good argument for using English master scripts as a ‘baseline’ standard for all versions of a piece of content. If you start from the same language for every translation, you can maintain a greater consistency regarding allusions to specific people, places or events. If a global production adopted a ‘source to target’ approach for some languages and used an English master script for others, then you’d lose this single source of truth that the master script currently provides.
Then there’s the fact that localization workflows to date have been largely based on having an English language source. We are still some way off accessing platforms that can support workflows in multiple languages to the same standard, although it will surely happen eventually – probably with AI as part of the mix.
Ah yes – AI. It is unlikely to have escaped your attention that AI is now routinely talked about as being the future of automated translation, subtitling and even dubbing. While it is true that the standard of AI-driven platforms is improving rapidly – to the extent that some are now claiming to achieve as much as 90% accuracy – that remaining 10% requires a skilled localization expert to refine the output. And of course, the more languages involved, the more localization experts are required. And these specialized localization skills are in short supply.
There are no easy answers here. Ultimately, translating content directly from the source to target language should improve efficiency and the quality of global content, but it will require a significant investment in the development of both human expertise and the underlying technology. The question remains whether it’s an investment the industry is willing to make.