Machine translation (MT) has improved greatly in the last few years -- to the point where its quality is of an acceptable level and fit for many basic purposes. However, as with all processes in which some form of artificial intelligence intervenes, the best MT output will only be obtained if a human intervenes -- as a post-editor -- to ensure that the target accurately reflects the source in terms of information content and to ensure that the text reads well.

The main advantage of post-edited MT (PEMT) is, obviously, the cost savings. The main disadvantage is that MT done using web-based tools places your text in the public domain, with the consequent issues of privacy and confidentiality.

During 2019 I have plans to develop a desktop-based MT tool that will overcome some of the confidentiality and privacy issues associated with using web-based tools.

In the meantime, interested clients may contact me to request web-based PEMT services, but will be asked to sign an agreement in which they accept the potential risks for their data.