Machine translation

Machine translation means translating text automatically from one language to another. It is usually used to make the translation process more efficient; the machine first produces a pre-translated version, which is then given to a translator for postediting. 

Machine translation can also be sufficient on its own if the translation does not have to be fully finished or if you just want to find out quickly what the text is about.

Lingsoft’s own machine translation technology

Our machine translation technology is based on open source machine translation technologies. We continuously develop our machine translation tools for different languages.

We can also create customer-specific machine translation tools that are customised for the customer’s own text content. The customised machine translation tool is trained using the customer’s own materials, which enables producing even better quality cost-effectively.

Postediting – the most efficient way to ensure high-quality machine translation

The most high-quality translations are still produced by skilled human translators. Postediting combines machine translation with the finishing touches of a skilled human translator. This allows us to provide our customers with high-quality translations quickly and cost-effectively.

The quality of machine translations has improved considerably in recent years as a result of neural network translation. Professional translators are especially needed for the translation of creative texts (transcreation), as people are better at processing and translating texts that contain metaphoric language and other types of language use that require creative interpretation.
Machine translation is at its best when translating concrete and unambiguous texts, such as reports, manuals, and instructions. 

However, postediting can also be useful in creative translation. For example, product descriptions for online shops may require creative translation, but they also contain many standard elements, such as price and VAT information. 
Lingsoft’s team will help you find the best combination of services for your content.

Lingsoft participated in the ELRC panel discussion on the state of language technology in Finland

The Language Bank of Finland organised an interesting workshop on the state of language technology in Finland. Lingsoft was invited to participate in an expert panel discussion, which included professors from the University of Helsinki and the University of Turku as well as representatives from other Finnish language technology companies. 

The members of the panel agreed that Finland has a strong tradition in the field of language technology and that there are good prerequisites for the development and maintenance of language technology in Finland. 

Lingsoft’s language technology development has a solid academic foundation, as it is based on a language structure analysis carried out by Kimmo Koskenniemi and Fred Karlsson at the University of Helsinki. Cooperation with universities remains an important part of Lingsoft’s operation.

Lingsoft’s neural network translations in the European Language Grid

In 2019, Lingsoft received funding from the European Language Grid for a project to integrate our language technology tools, including our neural network machine translation tools, into their platform.

During this project, our internal machine translation tool was upgraded into a modern SaaS solution based on Docker and SAPI technologies.

This means that customers can easily integrate Lingsoft’s neural network machine translation tools into their own solutions, for example, to support their own translators or customer service personnel or to implement multilingual archive searches.

The European Language Grid supported the LSDISCO project through an open call for pilot projects. The European Language Grid has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement № 825627 (ELG).

Contact us