Structuring of raw source manuscripts based on the requirements of the Client for the basis of copyediting and composing.
Pre-editing consists in processing the texts before machine translation. It typically involves correcting mistakes in the source text (mainly grammar, punctuation and spelling), removing ambiguities and simplifying structures. For statistical MT, it may also involve adapting the text in such a way that the input text is a closer match to the texts the engine has been trained with, which can help the MT engine perform better.
When a client sends us a scanned file or a very low quality image, sometimes upon converting it with typical text recognition software, certain parts remain illegible. That’s when the role of the pre-editor comes into play. What they have to do is read the original file and complete the parts that were left illegible in the converted file.
Among the stages of translation, pre-editing is one of the most interesting, as well as least known, as it helps salvage poorly-converted files and thus be able to process them with translation tools, whether in order to reduce costs or to create memories that we otherwise wouldn’t be able to.