Machine Translation Post Editing (MTPE) is a translation method that involves both humans and machines. A human, usually a linguist does the job of editing an automated translation with the sole purpose of refining the text and enhancing its quality. This hybrid translation ensures that translations sound more natural and the meaning of the primary text is retained.

MTPE is ideal for people looking to translate voluminous content within a short period and at a fair price without compromising on quality. It is a middle ground that boasts more accuracy than raw machine translation, while equally time-saving and less expensive than pure human translation.

Setting yourself up for success - MTPE best practices

MTPE, despite being relatively new in the world of translation, has quickly become one of the most preferred solutions to language translation. To ensure top-quality precision and accuracy, here are some best practices to follow.

• Ensure that the source text is error-free. Errors in a text can multiply and create problems as you go along. It is vital that human editors check through the source text before it is fed into the machine to certify that it is error-free, concise, and clear.
• Provide the machine translator with assistive data such as industry-specific terminologies, glossaries, and term bases. It is always easier to edit neural machine translations with data-driven outputs.
• Avoid over-stylistic edits. Sometimes the output of the MTPE might feel less natural when compared with a human translator. This is quite okay as much as it properly produces the original meaning.
• Use advanced Translation Management Systems (TMS) while editing. Using TMS improves the post-editing process. It helps save time and produces better consistency.
• Select the best translation engine. There are numerous machine translation engines to choose from and new ones are being developed regularly. Each engine performs differently. Some perform better than others, while some are more language- and domain-specific. Selecting the best engine saves you a lot of effort and time.
• Request for feedback from editors. Feedback comments such as most common mistakes and quality of machine translator can help in improving the accuracy of machine translation and making subsequent post-editing jobs easier.

There are two types of MTPE and they usually reflect the expected level of quality in a project

Light post-editing
This handles only the basic post-editing requirements. Human editors only adjust the output from the machine concerning spelling mistakes and semantics and check whether there is no wrong information. This type of MTPE is usually less expensive and effective for projects where quality is not a major concern.

Full post-editing
This focuses on effecting structural and stylistic changes. It is concerned with producing an output with a quality comparable to human translation. Although this type of MTPE is more expensive and time-consuming, it is usually error-free as it cuts across spelling and grammatical corrections, linguistic expressions, and writing styles. It is mostly preferred by people seeking quality.

Conclusion

Overall, MTPE is gradually replacing raw human translation as it can produce similar quality at a reduced cost and over a short time. This form of translation is shaping up to be the ideal solution for most industries and professional translation projects.