LVL goes to Hungary

Eydisphoto

In September Eydís will be representing our group at the CogInfoCom 2017 Conference. This will be LVL’s second time at CogInfoCom in Hungary, and we hope it will be an exciting conference with valuable insights. She will be presenting her paper “Cognitive workload classification using cardiovascular measures and dynamic features.” This will be Eydís’ last conference and paper before her study abroad semester starts later this year. So we hope she stays healthy and enjoys her time in Hungary.

 

Largest Icelandic LVL Group at Interspeech in 2017

This August, our LVL group members will be attending Interspeech 2017 to present their three papers, meet other folks in the speech recognition field, and have lots of fun. Two posters,“Building an ASR corpus using Althingi’s Parliamentary Speeches” and “Building ASR corpora using Eyra,” will be presented during the Wednesday Special Session: Digital Revolution for Under-resourced Languages 2 poster session at 13:30-15:30 so go say, “Hi!” to our members if you can. The third paper, “Objective Severity Assessment From Disordered Voice Using Estimated Glottal Airflow,” will be presented as a talk by Yu-Ren on Monday afternoon.

For the Alþingi Speech paper by Inga Rún, the language corpus can be found at Malfong.is and the Kaldi recipe can be found on Github.com but the best resource will be Inga Rún herself so grab a drink and find her at the Welcome Reception or the Standing Banquet.

We hope to meet you all there!

Article Publication in the Periodica Polytechnica Electrical Engineering and Computer Science journal

At CogInfoCom 2016, Eydís Huld Magnusdottir, gave a great presentation about Monitoring Cognitive Workload Using Vocal Tract and Voice Source Features. It was so informative in fact that the Periodica Polytechnica Electrical Engineering and Computer Science, a great open access journal, has published it. The article has the same name as the conference paper, Monitoring Cognitive Workload Using Vocal Tract and Voice Source Features. Eydís’ article is about the study she did in Iceland involving cognitive workload, Stroop tasks, and nearly 100 participants to find out whether vocal tract features perform better or worse than voice source features.

wave_karma
Figure: The results of speech formants tracking using the KARMA algorithm

 

To read the exciting conclusion, go directly to the journal’s website and read about it from home!

 

 

Paper accepted to an IEEE-Transactions journal

Congratulations to Yu-Ren for getting his paper accepted to the highly recognized scholarly journal in the field of speech processing, “IEEE/ACM Transactions on Audio, Speech, and Language Processing”! However, the paper wouldn’t have come about without the collaborative efforts of MIT Lincoln Laboratory, Universidad Tecnica Federico Santa Maria, Massachusetts General Hospital, and Harvard Medical School. His paper, “Evaluation of glottal inverse filtering algorithms using a physiologically based articulatory speech synthesizer,” is about measuring the performance of different algorithms on glottal flow generated by the VocalTractLab speech synthesizer. Since the paper just got accepted, we don’t have a link to it yet but we will add one as soon as it is available.

 

Edit: Volume 25 Issue 8 with Yu-ren’s article is now out. For those with a subscription, here is the link: http://ieeexplore.ieee.org/document/7946161/

post_body
Main results showing the performance of inverse filtering algorithms

Learning from Europe

Two of the LVL team, Jón and Anna, are working on a strategic plan for Icelandic language technology, together with Steinþór Steingrímsson from The Árni Magnússon Institute of Icelandic Studies. The three of them have been – and will be – meeting LT specialists in Europe to learn about successful technologies and to collect material for the strategic planning. February 15th to 17th they went from seeing Etienne Roth from EPC in Mannheim, Germany, on spell and grammar checking, to Luxembourg to speak with the machine translation team of MT@EC and over to London to meet speech synthesis specialists at Google. A lot of nice, clever and helpful people they met there! While digesting the mass-input of these three days, the preparations for the next trips are already ongoing: Estonia on the 1st of March and Gothenburg, Sweden, later the same month.

img_0227

Szymon Klocek, Markus Foti, Jón Guðnason, Steinþór Steingrímsson, Michael Jellinghaus and Andreas Eisele at EC in Luxembourg

img_0240

Jón at the Goodenough House, where he lived while studying in London “some” years ago

img_0258

View from the 11th floor, Jón and Steinþór together with Oddur Kjartansson at Google, London

 

The future of Icelandic

The event “Educational day 2017”, organised by SA-Business Iceland on February 2nd, focussed on the  Icelandic language and language technology. Leading persons from academia, industry, politics, and culture spoke with enthusiasm about the role and future of Icelandic and the chances within language technology. The president of Iceland, Mr. Guðni Th. Jóhannesson (whose name posed a challenge to the pronunciation talent of Amazon’s Alexa), also honoured the meeting and professed great optimism for the Icelandic language and LT. As did our representative from LVL, Jón had a talk on speech recognition, “Speaking Icelandic to computers”.

img_0188

ICASSP 2017 New Orleans – we will be there.

Two members of our lab will attend ICASSP 2017 conference in New Orleans. We will present a paper on the classification of voice modalities using a neck-surface accelerometer. The paper is the result of our collaboration with guys from the Massachusetts General Hospital, US and Universidad Tecnica Federico Santa Maria, Chile. Thanks for your help!

Hopefully, the trip won’t be all work and no fun.