Robert Eklund’s Linguistics Profile

Most of my linguistic research activities are described in more detail on the Research Page, but a short overview is provided here.

I studied General Linguistics, Speech Technology and Computational Linguistics at Stockholm University during the period 1990 through 1993, and was accepted as a PhD student in Computational Linguistics at the Institute of Linguistics there in 1993 (1995, formally).

I then became the first amanuensis (basically teaching position) in General Linguistics, a position I held for the (maximal amount allowed) three semesters. In 1994 I was employed as a consultant at Telia Research to help create the first concatenative speech synthesizer for Swedish.

Robert Eklund, Rapport Sommarmorgon, 31 August 1998

TV2 Rapport Sommarmorgon, 31 July 1998

After that, I was hired as a phonetician/linguist and worked for a number of years on automatic speech-to-speech translation, animated speech synthesis and (to a lesser extent) dialog systems, with an increasing degree of behavioral issues in human–machine interaction considerations.

In 1999, I moved my PhD studies (in Computational Linguistics) to Linköping University (since both Telia and Linköping University were formal partners of “Företagsforskarskolan” a collaboration between academia and industry) where I completed a thesis on “disfluency” (how, and why) you are not completely ”fluent”") in spontaneous speech.

Robert Eklund, TV3 Direkt, 3 November 1998

TV3 Direkt, 3 November 1998

During the same period) I also worked with my colleague Anders Lindström on “xenophones”, i.e. speech sounds not inherent to a given language, but are nevertheless used by speakers (like e.g. dental fricatives in Swedish).

For fun, I also performed (a minimal amount of) field work in Papua New Guinea, and published a couple papers on Tok Pisin, the pidgin/creole language spoken there.

During my time at Telia, I was responsible for media contacts (as evidenced by the two TV captures above and to the left), and appeared in the press on numerous occasions. I also gave hundreds of demonstrations of speech technology applications developed at our lab, as well as giving introductory talks on speech technology at most Swedish universities.

My last task at Telia was to help develop the first free-speech customer entrance to 90 200 customer service, where I collaborated with several people and helped develop the “prompt piano” to help collect ecologically valid speech data.

During 2005/2006 I was hired as a postgradual student at the International Computer Science Institute at Berkeley, California, where I worked on dialog (meeting) analysis.

Upon my return to Sweden in 2006 I commenced a second postdoc at Stockholm Brain Institute, (Karolinska Institute), Stockholm, performing fMRI studies on speech/disfluency perception.

Since March 2008 I work at Voice Provider, mainly as a dialog designer for automated dialog systems, but I am also involved in other activities like quality control, speech coaching etc, while pursuing neurocognitive studies at the Stockholm Brain Institite, now as an affiliated researcher.

Robert Eklund in Telia anechoic chamber

Although disfluency, for natural reasons, should be considered one of my “specialities” my main contributions – or “claims to fame – are my work on xenophones and ingressive speech, which is why these two areas have their own pages.

Also, one of the most linked-to pages of mine is (in all likelihood) my Tok Pisin page, found by pressing the Tok Pisin button above.

Robert Eklund The Lecturer

... no more!! This is the true story of an enigma resolved...

An anonymous ex-student of mine put this FANTASTICALLY adequate drawing of me in my pigeon hole. (The details, six arms, green skin and the exaggerated hair length, are due to a linguistic task I used to give my students.) I had never seen such a spot-on depiction of myself! For those of you who don’t know me, or haven’t endured my lectures, I hereby inform you that this is the very spitting image of me indeed!

I desperately tried to find out who the artist was, and spent hours trying to match the handwriting on the little post-it note that came with the picture with registration cards and assignments... but in vain. Ever since, the drawing has been hanging on the wall in my home, and for two years I had a “Wanted” page on my homepage in an attempt to track down the said artist. And... after two years, I received an email from her! (I invited her to a Thai restaurant in Stockholm, where we discussed The Canadian sci-fi series Lexx, among other things.)

Robert Eklund The Lecturer

Spitting image rendering of Robert Eklund lecturing