Many thanks!

We have received your enquiry and will be in touch as soon as possible.

SwissGlobal Newsletter

Lost in translation? Not with SwissGlobal. Stay up to date with language industry news, tips, interviews and more – subscribe to our monthly newsletter.

What does a computational linguist do?

Technology and linguistics are two sectors that have become more and more intertwined in recent years – and can now be learned as a combined discipline. To meet the constantly changing demands, there is of course a need for qualified professionals: experts who drive new innovations and solutions on both a linguistic and technological level.

One of them is Alla Stöckli. Trained as a linguist and language teacher, she completed her Master’s degree in Multilingual Text Analysis (MLTA) / Computational Linguistics at the University of Zurich in 2021.

Today, she works as a Solutions Architect at SwissGlobal. Here are her insights into the work of computational linguistics.

Alla, you work as a Solutions Architect in the field of computational linguistics. What does this term mean in your own words and what do computational linguists deal with?

Computational linguistics is a very diverse field with a myriad of possibilities. In essence, a computational linguist is someone who works at the intersection of technology and linguistics, using technology to process natural human language.

In practice, this can include many different tasks and activities. There are various areas within Natural Language Processing (NLP): machine translation (e.g. Google Translate and DeepL), speech recognition and production (e.g. voice assistants such as Siri and Alexa), etc. Computational linguists are involved in such projects to facilitate the understanding of natural human language by a machine.

Only 10 years ago, this area was still in its infancy. Where are we today?

Nowadays, everything in our life becomes automated. Although the concepts of language processing are not new, the technological advancements of recent years, such as deep learning, have had an immense influence on the development of NLP. This development has made many things in NLP possible and has made it a very attractive field to work in.

The language and technology sector is constantly evolving at a rapid pace. One might ask if information is still relevant at the end of one’s education when pursuing a degree in this field. What are your thoughts on this?

In my experience, everything that one learns during studies can be applied in practice. The skills acquired during studies always stay relevant, only their application changes somewhat. Moreover, even if certain topics or tools become obsolete, they still serve as a basis for new knowledge or new practices. The important thing is to stay informed about the new developments and not to get stuck in old thinking patterns.

Your academic and professional background is in linguistics. Later, you’ve immersed yourself in the technology sector and learned how the two go together. What prerequisites should someone have who wants to work in computational linguistics?

Depending on the area of NLP one would like to work in, the job of a computational linguist can be connected to more or less technology. But overall, I find the following points to be important:

  • Deep understanding of language. Systems are very language specific. There is no one-fits-all solution, and even if there was, it would not be really good. A passion for language and an understanding thereof are key to successful work in NLP. Depending on the field, it may require in-depth knowledge in specific disciplines. For instance, speech processing and production require a deeper understanding of the phonological system of a language.
  • Programming. The most popular programming language for NLP is Python. It has many useful libraries that are used for language processing.
  • Analytical and solution-oriented thinking. An analytical mindset definitely helps to break down linguistic concepts and organise them in a way that is more understandable for a machine. Solution orientation is another important aspect, since working with machines and developing models usually entail a certain amount of frustration and failure. To overcome problems, one should think in a solution-oriented way.
  • Willingness to learn new things. As already mentioned, everything continues to develop. The same should be true for a computational linguist’s knowledge and skills.

Do you have predictions for the field of computational linguistics? What are the trends and needs? What skills will be needed in the future?

The world constantly comes up with new technologies, which makes it easy to lose the overview. This is why in the future businesses will need people who have expertise in technology and the know-how to integrate it into business processes for it to generate value. Such people are solutions architects and digital transformation enablers who know how to squeeze the maximum value out of tools, apps and models. In my opinion, the need for such people will increase, also in the field of NLP.

As far as the development in the computational linguistics are concerned, there are some encouraging trends. NLP resources are becoming more accessible and the NLP world itself more solidary, as various companies are making their algorithms/models/software available as open source. Furthermore, a lot of resources are put into the digitization of the so-called “low-resource languages”, for which not so much training data is available. This is also true for the Swiss German dialects. Hopefully, we will soon see some progress in this area.

Alla Stöckli has been working as a translator and language teacher since 2014. She’s completed her Master’s degree in Multilingual Text Analysis at the University of Zurich. For those interested in a career path in Computational Linguistics: you can find more information here.