Machine translation (MT) post-editing is quite simply the editing by humans of the content translated through MT engines. As the demand for translation of huge volumes of content at a previously unimaginable scale increases, the demand for MT-ed content has increased. This, in turn, has driven up the need for post-edited MT content. Both acronyms MTPE as well as PEMT are used to describe this service.
Get quoteAs both the terms are little obvious to understand; Machine translation refers to the translation performed using a translation machine engine, whereas the traditional way of translation is done by using human language experts. To make it easy for you to understand, few of the major parameters to differentiate it can be: a. Quality of output – Human translation preferred over Machine translation; b. Suitable for business – Human translation preferred over Machine translation; c. Turn-around time – Machine translation preferred over Human translation; and d. Cost effective – Machine translation preferred over Human translation. Considering the PROs and CONs, LSPs like Braahmam also recommends a hybrid approach for translations, a mix of Machine translation and Human translation widely known as Machine translation post-editing.
In general, there are two levels of post-editing your LSP will offer: light and heavy or full. Light post-editing is used when the raw output is not of very poor quality or when “good enough” quality will do. Generally, the types of errors that need to be corrected in this approach are lexical and syntactical. In heavy or full post-editing, the goal is to equal human-quality translation. As such, apart from the above-mentioned error types – lexical and syntax – style, fluency, and less obvious errors too are corrected. To make post-editing fruitful, sit down with your language service provider (LSP) providing the service so that everyone is on the same page. Your LSP can also provide you structured feedback which will help you understand the types of errors creeping up in the output and go towards improving the engine over time.
Please write it to sales@braaham.net and we will get back to you.
Machine translation brings speed, it can chew through gigabytes of content, and because of these two reasons it makes translation possible for many content types that would never see the light of day in another language otherwise, and so is preferred by marketeers and global content managers. However, MT content is still not on par with human-quality translation in many languages and domains, even with the newest MT kid on the block: neural MT (NMT), but industry leaders keeps on investing to confirm/ verify, how close MT strings are to human translation (BLEU score).Industries such as legal, travel, IT/software (knowledge bases, user guides, manuals) automobile, education and training are the usual candidates for MTPE.
Yes, likewise there are few guidelines to get the maximum value out of MT post-editing. Few are: a. First off, write for machine translation – Follow best practices while creating source content that helps in getting relatively error-free raw MT output and thus bringing down the post-editing effort; b. Pre-edit where required – Use some tweaks such as fixing spelling errors, formatting the document, and flagging words not to be translated will contribute to a better output; c. Invest in trained MT engines – Put your time, effort and money in modeling a custom engine suited to your domain and target languages; d. Work closely with post-editors – Work with language-cum-domain experts, set expectations from the content and they will help you arrive at the quality that will align with these expectations; e. Choose the level of post-editing carefully – No hard and fast rule here, analyse every content project and determine how much editing you require i.e. light and full post-editing. Post-editing gives the control you seek and lets you smartly leverage MT to the maximum.