- What is MTQE?
- Why are MTQE algorithms important in translation?
- Examples of MTQE applications
- The role of MTQE in translation
MTQE algorithms have transformed the way we assess the quality of machine translation. In an earlier article, we discussed metrics such as TER and BLEU, which also serve this purpose but rely on reference texts. But what happens if no such reference exists? This is where MTQE comes in, offering unique possibilities and real value to the translation industry.
What is MTQE?
MTQE, or Machine Translation Quality Estimation, is an advanced set of algorithms designed to evaluate the quality of machine translations. What makes it different from TER and BLEU is that it does not require comparison with a reference translation. Instead, MTQE operates on a different principle, using:
- linguistic feature analysis (such as grammar correctness and terminological consistency)
- machine learning techniques
- statistical indicators
- contextual elements
MTQE not only determines the overall quality of a translation but also pinpoints which parts of the text need a translator’s intervention. The algorithms learn to recognise patterns of both good and poor-quality translations by analysing vast linguistic datasets.
Why are MTQE algorithms important in translation?
The use of MTQE algorithms brings significant benefits for translators. It is a step forward compared to relying solely on TER and BLEU in quality assessment. MTQE is one of the tools that enhance the modern translator’s toolkit. But what exactly can it improve in practice?
MTQE and time efficiency
MTQE saves a considerable amount of time. Before it was developed, machine translation quality could only be measured against a human-produced reference translation, which was highly time-consuming in large projects. MTQE removes the need for this step, allowing instant evaluation of machine-generated content.
MTQE and translation quality assessment speed
MTQE works like a quality sensor. It immediately signals whether a passage needs more thorough editing. This helps both translation agencies and freelancers plan project timelines more effectively.
It also optimises the overall process. By precisely evaluating the quality of each segment, MTQE directs translators’ attention to the parts that need the most work.
Examples of MTQE applications
The potential of MTQE algorithms is impressive and their role in the industry cannot be overstated. They are especially valuable in large-scale projects, where quick quality assessment helps select the best translation tools.
Managing large translation projects with MTQE
Take a large technical documentation project spanning hundreds of pages. Even with a sizeable team, preparing a reference translation would be a challenge. With MTQE, every segment of the machine-translated text is automatically analysed and marked for human review where needed.
This makes project planning more efficient and significantly accelerates delivery, which has a direct impact on client satisfaction.
Testing translation engines with MTQE
MTQE also supports A/B testing of different machine translation engines. A translation agency or freelancer can easily compare outputs from different providers without creating separate reference translations.
For example, the same text can be translated with DeepL, Google Translate and Microsoft Translator. MTQE-based tools then evaluate the results, helping to identify the engine best suited for a specific type of content. This leads to higher quality final translations.
The role of MTQE in translation
MTQE represents a breakthrough in assessing machine translation quality. It allows fast and accurate analysis of translated text without relying on reference material. For companies and freelancers, this means time savings, streamlined workflows and the ability to choose the most effective MT engine. Its benefits are comparable to automation in translation agencies.
MTQE algorithms continue to improve, delivering increasingly reliable quality evaluations. If you use CAT tools, LivoTRANSLATE, translation memories or glossaries in your work, MTQE will complement your workflow by helping assess the performance of MT engines. These are tools for businesses and freelancers – for those who use machine translation when translating texts.
