A Q&A With Maria Castañeda of Lokalise

Arabizationa cloud-based localization and translation management system, has noticed a fascinating shift in how content is translated and has some interesting new data on the rise of machine translations versus human translations.

Their findings show a significant increase in machine translations in 2024. I asked Maria CastanedaSenior Product Marketing Manager at Localise, asked us a few questions to provide further insight into these findings.

What are some of the main differences between machine and human translations in terms of quality and efficiency?

Machine translation methods vary greatly depending on the technology used. Using Lokalise’s automated methods, such as Lokalise AI, we achieve 80% quality that our customers repeatedly rate as comparable to human translations. These methods offer unparalleled speed – delivering translations up to 10 times faster – and efficiency, eliminating much of the back-and-forth typically involved in traditional processes. This efficiency translates into an 80% reduction in the cost of translations, making automated solutions not only faster, but also more affordable.

While previous automated approaches struggled with context-specific translations, recent developments such as Lokalise AI effectively address this challenge. By leveraging glossaries, style guides, and keyword context, Lokalise AI ensures more accurate and consistent translations, closely matching the original intent and style.

However, human translators still provide higher quality for certain scenarios, especially when dealing with highly sensitive data or highly accurate translations. In these cases, automated methods often leverage a human-in-the-loop approach to review and improve.

At Lokalise, we value the evolving role of humans in the localization process. With current developments, their role can shift from pure translation to reviewing and dealing with precise or specialized translations, improving workflow and further enhancing quality. This synergy between human expertise and automation ensures the best results for diverse translation needs.

Are there any other insights into what caused this sharp decline in both machine and human translations in late 2023 and the sudden spike in machine translations in late 2024?

The sharp decline in both machine and human translations in late 2023 can be attributed primarily to the usual seasonal downturn associated with the holiday period. This time of year often sees a slowdown in business operations as companies and individuals focus on holiday activities, resulting in lower demand for translation services. In addition, this period may coincide with the end of budget cycles.

The significant increase in machine translations in Q3 2024 can be explained by several key factors. First, there is a noticeable trend for companies to increasingly adopt digital transformation strategies. As companies seek to expand their operations globally, machine translation tools have become more attractive due to their cost-effectiveness, speed, and significant advances in accuracy and contextual relevance over the past few years.

Moreover, late 2024 may see many companies joining directly into machine translation services as part of their strategic initiatives to streamline operations and integrate new technologies. This transformation could have been driven further by advances in artificial intelligence and machine learning.

Which languages ​​and/or regions have seen the most measurable improvement through machine content translations, and why?

The improvements brought by machine translations of content have been global, with no particular language or region seeing disproportionate growth compared to others. This global boost stems from advances in artificial intelligence and natural language processing (NLP), which have improved translation quality across a wide range of languages, including languages ​​with complex grammar or structure.

For product launches and updates, machine translation tools have made it easier to adapt content to suit a global audience, ensuring companies can communicate effectively in multiple markets. This global application demonstrates the scalability and effectiveness of automation in driving international growth, regardless of the language or region in question.

Related articles

All the news that fits FAST

Broadcast news is one of the most exciting and rapidly growing areas of development for FAST, which has been highlighted by the recent announcement of the BBC/AMC’s 24-hour live news channel FAST. Older media news broadcasters, such as Scripps, Sinclair, Hearst and USA Today, have also entered the FAST news space, many of them focusing on local content. What are some current trends and challenges for delivering news quickly? In this article, several leading industry commentators provide insight.

May 14, 2024

Localization and the future of the flowing ecosystem

Broadcast localization is not just about delivering content to multiple regions; It’s about “opening up the world,” says Quincy Olatunde, VP of Product at Peacock TV, who explains why he thinks streaming localization is crucial in this clip from his keynote conversation at Streaming Media Connect 2023 with Media Universe cartographer Evan Shapiro.

September 27, 2023

How to make subtitled streaming work

While offering local content on streaming platforms still faces logistical and legal challenges, there are new ways to deliver a regional user experience via apps that can feature hyper-local channels that can help bind communities together.

August 22, 2022

National OTT services and translated content

Evoca’s Dana McLeod and USA Today/Gannett’s Carolyn Harris discuss the benefits of hyper-localization of OTT content for both major providers (who appear to be moving in the opposite direction) and end users (who are demanding it) in this clip from their panel at Streaming Media West Connect 2021.

November 17, 2021