Why artisan and at-scale localization services aren’t mutually exclusive

29th November 2021

Next Post

We work in a creative industry.  So much effort goes into creating content of the highest standard.  Production teams finesse every word of their scripts, carefully cast actors and voice talent, handpick the best crew, and then meticulously shoot and post-produce the content until it’s as close to perfect as possible.  Then that beautiful work of art is handed over to localization and access teams to reversion it for multiple international audiences at a low cost, in record time.  The result is that only a fraction of your audience gets the experience you intended, and your content is massively compromised on a global scale.   

It’s an understandable situation. The digital revolution and subsequent explosion of streaming services mean there’s suddenly an unprecedented demand for global and accessible content.   With thousands of subtitling and dubbing units per production, we simply haven’t had the time to develop and grow the resources needed to match the demand. In this situation, it’s easy to believe that we have no option but to sacrifice the quality of these services to make up the quantity.  But there are ways to improve the efficiency of our localization and access service workflows so that our teams have enough time to maintain – and even enhance – the creative integrity of the original production.   

Imagine that your localization and access teams are an orchestra and each of the talented experts translating, dubbing and subtitling the content are musicians playing different instruments.  In the same way that these musicians are brought together by a single piece of sheet music to create harmony, so the transcript could be the unifying factor in localization and access services.  At the moment, we create various transcripts for different stages of these processes – potentially introducing different problems at each stage and resulting in various members of our localization orchestra working with slightly different sheet music.  If we focus instead on creating a flawless master transcript that serves as a blueprint for all the localization and access service processes and provide this transcript as interchangeable data that can be repurposed to fit the different requirements of each process (eg: proving dubbing artists with speaker labels and subtitlers with character per line information) then our orchestra would spend less time correcting their sheet music and have more time to focus on their performance. 

By taking away the pressure of technical accuracy and providing one “source of truth” the transcript supports a more artistic process.  When our teams aren’t worried about basic mistakes which can have significant implications on comprehension – like hour vs our and plane vs plain – they can focus instead on ensuring that we translate and communicate not only the words but the meaning behind complex ideas and idiomatic expressions like “raining cats and dogs” and “fish out of water.”

There are other ways we can unlock opportunities to scale without compromising on the quality of our localization and access services.  Adopting cloud-based tools and platforms helps speed up workflows – from using camera-to-cloud technology to share rushes directly from location to using virtual storage for easy file sharing and media management – the cloud helps us tighten up timelines so that we’ve got more time available for creative work.  Similarly, automation tools help us reduce how much time is spent manually managing workflows and processes.  Standardizing the way we work by adopting an agreed set of formats and templates across the industry would also reduce the amount of time spent repurposing what is largely the same set of information into multiple deliverable formats for global distribution.     

Obviously, artificial intelligence has a major role to play, but not necessarily in the way we have been led to believe.  Rather than using ASR and online translation tools as catch-all problem solvers to replace the roles that people play in access and localization services, we should think of AI as a tool to provide people with more time to do high-value work.  The role of AI could be as simple as using spell-check, capturing text on screen with optical character recognition tools or using time alignment to line up subtitle text with the corresponding audio.  If AI can do 90% of the low skill, heavy lifting then your team will have a lot more time for creativity.  After all, there’s nothing new about using computers to help us do our jobs more efficiently.  

Finally, by treating localization and access services like a closely guarded secret we make it impossible to scale. The artists that make up localization and access teams are critical to the creative process, but we need to rethink the roles of transcribers, translators and quality controllers.  Our teams need to be multiskilled, with translators that understand how to subtitle and that are equally adept at spelling and grammar as they are fluent in languages and captioners that are as comfortable with origination as they are with quality control.  When teams understand the full process that they’re contributing to, and are familiar with the content they’re working on, then they become both more efficient and more creative.   

We have a responsibility to our clients, and to the original production, to provide artisan services, even with our limited resources and timescales.  There’s no question that we need to create efficiencies to meet the demand we’re facing, but we should look to create these time-savers in other areas so that we can protect the creative process and give our audiences the video experiences they deserve. 

This article was originally published in the 2021 Fall issue of the EGA Reller Magazine   

Contact the Take 1 team to find out more.