Google Translate is comparatively the easiest to use for the end-user, as it is available via the internet. The engine has analysed a large number of texts over several years, during which the neural-networks-based machine has learned a huge amount of information. This makes Google Translate a very powerful and accurate machine translator. Its biggest benefits, from a service point of view, are cost savings and ease of use. However, one of the biggest disadvantages is the public nature of the service, as it does not enable GDPR-compliant data processing for sensitive information.
GDPR-compliant data management is emphasised in the Amazon Translate service, where data are stored in the AWS cloud, making GDPR-compliant data management and storage possible. The benefits of Amazon Translate are using the same neural networks method for translation as Google Translate and the additional possibility to store and manage data. The user can influence the development of the service itself, which makes the service better for certain domain areas compared to Google Translate. AWS knowledge in using Amazon Translate is vital, which can be seen as both a disadvantage and an advantage. The service can be expanded and taught, which makes the translation service more efficient, but this causes an increase in cost and requirements in AWS knowledge. The service costs are based on the use of Amazon Translate and its sub-services (e.g., ACT).
The European Commission’s eTranslation is the safe and free option. Compared to Amazon Translate and Google Translate, eTranslation has large sets of information from specific domains input to it – securing rather high-quality translations in certain domain fields. It can translate documents into several languages with one translation request. The challenges of eTranslation are related to its limited use: the service requires creation of EU Login credentials for each service user, and single words and texts under 30 words cannot be translated. eTranslation is also not capable of translating codes into text.
In order to make a more concrete and detailed analysis on the usability of machine translation services in different real-life environments, there is a need for more detailed information gathering and discussions with service providers regarding what is possible. The machine translator services’ capabilities and quality of translations in different fields of expertise also need to be tested with actual texts and documents, and with linguistic professionals to evaluate the accuracy of the translations. In addition, it would be of high interest to study the capabilities of machine translators in utilising specialised terminology and vocabularies. These specialised vocabularies have been created in recent years in different administrative branches for divergent use. It would seem highly reasonable to examine further whether and how these vocabularies could be used as an external source for broadening machine translator services’ vocabularies and utilise the excessive terminology work that has already been done.