AT LANGUAGEWORLD

An interview with Joel Escudé, our computational linguist

This week we interviewed Joel Escudé, our computational linguist. Joel studied Technical Engineering and specialised in Industrial Chemistry. To further his training, he completed a Master's Degree in Applied Mathematics at UPC. After finishing his final project on machine translation, he chose AT as the place to apply his knowledge of machine translation engines.

 

  1. How long have you been working here?

This March marks his one-year anniversary.

 

  1. How many languages do you speak?

I speak three languages: Spanish, Catalan and English.

 

  1. What was the last translated book you read?

Dream Story by Arthur Schnitzler, translated from German into English.

 

  1. Do you watch films in their original or dubbed versions?

If they are in a language I speak, I watch them in their original version. Otherwise, if a film is in a language I don't know, I like to watch it in the original version with subtitles. It has to be in the original version, so I can see the author's interpretation in its entirety.

 

  1. Do you like what you do? Why?

Yes, I love it. Here we research and develop machine translation and natural language processing. Each week brings a new challenge, as we come across different problems that test our ability to solve the technical aspects.

 

  1. In what direction do you think the translation sector is heading?

The translation sector is heading towards automatic translation and natural language processing tools, be it in terms of terminology, glossaries or some other aspect. This method of translation assists the translator, although the professional makes the final decision on the content of a translation or an edit.

 

  1. Has the way tools are used changed much since you've been here?

Yes. There are many tools still being used, although, whenever we believe a change could bring about an improvement, we test it.

 

  1. What is your favourite AT tool and why?

I haven't got a favourite tool specifically. What I can say is that, when editing any type of format, where we need to extract data from documents for analysis, the management process carried out using our tool makes it very easy for us to do our work.

 

  1. Of all the projects, which has been your favourite?

Right now, we are developing a specific machine translation engine for a single customer, and it is very interesting because of the technical challenges involved. It is a highly ambitious challenge, which is why it is one of my favourites. We are developing an engine that addresses the specific needs of this customer, so that the quality of their translation is the best possible.

 

  1. What is the most difficult challenge you've faced since you've been working here?

The biggest challenge has been harnessing our R&D resources, which are occupied with several projects at the same time: creating engines, natural language processing tools, customer-specific engines, etc. This involves using our data, computing resources and time as effectively as possible to constantly improve while testing new technologies that we could apply to our processes. We want to make the most of the knowledge we gain from all our experiments to bring about increasingly ambitious results more efficiently.

 

  1. What would you say has been your biggest personal achievement at AT?

My greatest personal achievement is related to the previous question. Overcoming the challenges we face at AT is an achievement for me. We are very ambitious at this company, and we always look to try new things to see if we can improve in quality, time, etc. We set these challenges for ourselves, and success is being able to achieve them together.

 

  1. Is there anything we do today that you think is no longer relevant?

No. I think that everything we do, whether it involves older or more innovative methods, is relevant. For example, if a previous technology gives us good results, it will continue to be important; we will continue using it until we validate a real improvement, for example, in a machine translation engine.

 

  1. In your opinion, what makes AT LS stand out from the competition?

The development of in-house technology and customer-specific translation services. This combination is what makes AT stand out from the competition. Also, by advancing its internal processes and its human team to deal with technical problems, it creates an added value in the translation industry.

 

  1. Why do you think translation is important for human communication?

It's very important if we don't speak the same language. Because we live in a globalised world where we can communicate from all over, translation is necessary.

 

Thanks to translation, our message has greater reach. Although Artificial Intelligence tools automate the process, it is always recommended that a translation professional review the content. New technologies have helped to streamline the translator's work and bring about increasingly high-quality results.