Machine translation (MT) is a sub-field of Computational Linguistics that investigates the use of software to translate a text from one natural language to another. Recent advances in deep learning have led to the rise of Neural Machine Translation (NTM) that applies a large artificial network to predict the likelihood of a sequence of words and even full sentences. This development has resulted in some hype over the MT quality, with companies such as Google and Microsoft, claiming that they reached human parity in some language pairs.
Although MT has still a long way to go until it is seen as a mainstream tool, organizations, companies and even language service providers have tried to use it in the last couple of years alongside translation memories and terminology databases to meet the high demands for content localization. While MT produces good enough translations when dealing with highly repetitive and specialised content, the quality is rather low when translating creative content where cultural knowledge is needed. Post-editors are employed to improve the quality of the MT output and prepare the content for publication and dissemination. This trend has led to the rise of new business models and hot debates in the industry about ROI, post-editing prices and the impact on the translator’s profession.
What Machine Translation capabilities does Wordbee provide?
In Wordbee you can use Machine Translation in two ways:
Enable MT Hive for your customers
Wordbee MT Hive is a module that enables customers to request an instant translation via the Client Portal. Workflows can be automatically kicked off in order to reduce management tasks and save costs.
If the instant translation meets the quality expectations, the customer just needs to download the bilingual files. If the translation needs improvements, the customer can submit a request for post-editing and launch the workflow. See the articles below to learn how to work with MT Hive:
Connect to commercial MT engines
Wordbee Translator integrates with 12 commercial MT engines and it has recently added the Intento MT connector that gives one-stop access to a wide range of third-party machine translation systems. Check the articles below to learn how to configure and enable an MT engine.
After you have enabled the MT system you would like to use in your project, check the articles below to learn how to add the MT to your workflow. Depending on your access rights you will be able to use MT to handle full documents or only segments in the Translation Editor. Also, if projects have been set up to fully automate the use of MT (when launched via the Client portal, for example), you will then be able to trigger alternative workflows.
NEW
As of October 2020, Wordbee Translator has added the Intento MT connector that gives one-stop access to a wide range of third-party machine translation systems. Intento provides a single interface to all those vendors and models. This way, Intento clients may evaluate all of them for their specific use. To sign-up, contact Intento at https://inten.to/. Also check-out this How-to article: How to machine translate with Intento APILearn more
If you are interested in implementing MT Post-editing in your localization workflow and you do not know where to start, here are some useful reference articles:
Wordbee blog articles on Machine Translation and Post-editing