LANGUAGE:
LANGUAGE:
Our people are our pride, helping companies resonate with their customers for 20+ years.
We create, transform, test, and train more content than anyone in the world – from text, voice, audio, video, to structured & unstructured data.
Solutions
Our experts know the in-and-outs of your industry & its challenges.
Harness the Lionbridge Lainguage Cloud to support your end-to-end localization and content lifecycle
Our people are our pride, helping companies resonate with their customers for 20+ years.
About Us
Key Facts
Leadership
Insights
News
Trust Center
We create, transform, test, and train more content than anyone in the world – from text, voice, audio, video, to structured & unstructured data.
Content Services
Technical Writing
Training & eLearning
Financial Reports
Multicultural Marketing
Digital Experience Assessments
Translation Services
Translation
Video Localization
Software Localization
Website Localization
Translation for Regulated Companies
Interpretation
Multilingual SEO & Digital Marketing
Content Optimization
Testing Services
Functional QA & Testing
Compatibility Testing
Interoperability Testing
Performance Testing
Accessibility Testing
UX/CX Testing
Solutions
Digital Marketing
Live Events
Machine Translation
Our Knowledge Hubs
Positive Patient Outcomes
Future of Localization
Innovation to Immunity
COVID-19 Resource Center
Disruption Series
Patient Engagement
Lionbridge Insights
Our experts know the in-and-outs of your industry & its challenges.
Life Sciences
Banking & Finance
Retail
E-Commerce
Games
Automotive
Consumer Packaged Goods
Technology
Industrial Manufacturing
Legal Services
Travel & Hospitality
Harness the Lionbridge Lainguage Cloud to support your end-to-end localization and content lifecycle
Lainguage Cloud™ Platform
Connectivity
Translation Community
Workflow Technology
Smairt Content™
Smairt MT™
Smairt Data™
Language Quality
Analytics
SELECT LANGUAGE:
Machine Translation (MT) has been around for decades. In recent years, it has evolved at an exponential rate. As companies produce ever-growing amounts of content in different languages, they have come to see MT as an opportunity to extend their reach in an increasingly globalized world.
Companies seeking to deploy MT can explore the following four basic strategies.
No matter which strategy you are contemplating, selecting the right engine may be challenging without the proper data and MT experience. Lionbridge is an expert in MT. In addition to having more than two decades of MT experience, we have gathered a large volume of linguistic and quality data about MT technology that will help you make the right choice. This webpage provides basic information about the performance of popular MT engines for the most common language pairs to help you select the best option based on your content.
When choosing among the many available MT systems, it’s important to note some engines address a specific function or domain. If your needs don’t align with that purpose, the engine may perform sub-optimally no matter how advanced it is. To determine the best option, first, identify why you are using MT.
If you want an MT engine for general use, it may be appropriate to use Google Translate or Bing Translator. If you seek MT services for a specific language or domain, you may achieve better results by turning to Amazon Translate or DeepL Translator.
Lionbridge’s Machine Translation Tracker analyzes engine performance monthly to help you figure out the best MT engine depending on the language pairs you use. The next time you ask which MT engine is best, reframe the question to, “Which MT engine is best for me?” And count on Lionbridge for guidance.
Want to learn more about the different types of MT technologies? Check out our blog Machine Translation in Translation.
Generative Artificial Intelligence (AI) has achieved a significant milestone: It outperformed a Neural Machine Translation (MT) engine in one of our comparative evaluations. Specifically, Large Language Model (LLM) GPT-4 provided slightly better quality than Yandex for the English-to-Chinese language pair, as shown in Figure 1.
This development is noteworthy because it’s the first time a different type of MT approach has beaten a Neural MT engine since the advent of Neural MT. Moreover, a non-MT approach — a multi-purpose language automation not specifically prepared for Machine Translation — has beaten the Neural MT engine.
Why should you care about this occurrence? If you are an MT provider, you must be at the forefront of technological advancements and consider how they will impact your current MT offering to stay competitive. If you are an MT buyer, you must be privy to these developments to make sound MT investments, which will likely include some LLM-based technology instead of pure Neural MT offerings.
It's worth noting that generative AI is still in its early stages. As such, it falls short in some key areas. For instance, it produces variable outputs during multiple runs, has Application Programming Interface (API) instability, and makes more errors than Neural MT engines. These issues must be resolved for the technology to mature, and we are already seeing improvements being made at breathtaking speed.
The incredible speed at which LLMs can improve supports the notion that LLMs will become the next paradigm for Machine Translation. We expect a hybrid period whereby Neural MT providers integrate some aspects of LLMs into the Neural MT architecture as the paradigm evolves.
Read our blog for a translation quality comparison between Neural MT and LLMs for two more language pairs and additional thoughts on whether it is the beginning of the end of the Neural Machine Translation paradigm.
—Rafa Moral, Lionbridge Vice President, Innovation
Generic Machine Translation (MT) engines will frequently provide an adequate output for companies seeking to automate their translations. However, these engines may produce subpar suggestions — especially when dealing with technological or highly specialized content.
Companies seeking to improve Machine Translation (MT) results to meet specific goals can consider two options: MT customization and/or MT training. Either method — or a combination of both — can produce better results during the automated translation process.
However, the approaches differ from one another, and they are not interchangeable. Table 1 provides an overview of MT customization and MT training and offers some considerations when evaluating each method.
MT Customization | MT Training | |
---|---|---|
What it is and how it works | An adaptation of a pre-existing Machine Translation engine with a glossary and Do Not Translate (DNT) list to improve the accuracy of machine-generated translations | The building and training of an MT engine by using extensive bilingual data from corpora and Translation Memories (TMs) to improve the accuracy of machine-generated translations |
What it does | Improves MT’s suggestions for more accurate output and reduces the need for post-editing | Improves MT’s suggestions for more accurate output and reduces the need for post-editing |
Specific benefits | Enables companies to adhere to their brand name and terminology and achieve regional variations | Enables companies to attain a specific brand voice, tone, and style and achieve regional variations |
The risks of using it | The MT could make poor suggestions and negatively impact overall quality when executed improperly | MT training may fail to impact output if there is not enough quality data to train the engine; the MT could generate poor suggestions and negatively impact overall quality if inexperienced authors overuse terminology |
When to use it | Ideal for technological and detail-oriented content and any content that requires: *Accurate translations of terminology *Regional variation, but you lack sufficient data for MT training |
Ideal for highly specialized content, marketing and creative content, and any content that requires: *A specific brand voice, tone, or style *Regional variation, and you have enough data for MT training |
Success factors | An experienced MT expert who can successfully manage input and output normalization rules, glossaries, and DNT | A minimum of 15K unique segments to adequately train the engine |
Cost considerations | There is a one-time cost to update the profile that goes into the MT engine and some ongoing costs to maintain a glossary over time; costs are relatively inexpensive when factoring in the potential benefits and are typically lower than MT training costs | There are costs associated with the first training and potential costs for additional training, which may be considered over time if the MT performance monitoring indicates room for improvement; MT training can be worth the investment in certain cases when factoring in the potential benefits |
Table 1. A comparison between MT customization and MT training
Read our blog to learn more about Machine Translation customization vs. Machine Translation training.
—Thomas McCarthy, Lionbridge MT Business Analyst
For more insights and future trends about Machine Translation, read our Future of Language Tech blog post – Future of Machine Translation.
Disclaimer