Baltic_HLT_Logo_BlueWhite_half.png
 

The 8th Conference

Human Language Technologies -

The Baltic Perspective

September 27 - 29 / Tartu, Estonia

REGISTRATION IS CLOSED

CONFERENCE

The 8th International Conference “Human Language Technologies - the Baltic Perspective” will take place in Tartu, Estonia, September 27th till 29th, 2018. Its deepest and darkest desire is to promote collaboration and exchange of ideas between researchers, developers, students and users of human language processing technologies. The main topic of interest is the state of human language technologies for the languages

spoken in the Baltic states.

 

Previously this conference has been held in Riga (2004, 2010, 2016), Kaunas (2007, 2014),

Tartu (2012), and Tallinn (2005).

 

This year the program of the conference will include keynote talks, a workshop, oral and poster presentations of your research papers, a conference dinner and social event, sponsor presentations + coffee and cookies! See details on the organizers, or check below for further details, important dates and submission information.

DSC_7870_edited
DSC_7870_edited

press to zoom
DSC_7783_edited
DSC_7783_edited

press to zoom
DSC_7851_edited
DSC_7851_edited

press to zoom
DSC_7870_edited
DSC_7870_edited

press to zoom
1/8

WORKSHOP PROGRAMME

IMPORTANT

DATES

MAY 21 Submission of extended abstracts

JUNE 20 Notification of acceptance

JULY 22 Camera-ready papers

SEPTEMBER 5 Demo abstract submission deadline

SEPTEMBER 27 Workshop and welcome reception

SEPTEMBER 28-29 Conference

WORKSHOP

We are happy to introduce the workshop "Wikipedia meets NLP" on September 27th

as part of this year's conference!

 

The workshop is organised by the Miljon+ initiative and will focus on Wikipedia as an open language resource. We will address topics such as Wikidata, linking Wikipedia with other ontologies and resources and automatic generation of Wikipedia articles. The workshop will include an invited speech and a hands-on session.

 
 

CALL

FOR DEMOS

In addition to research talks/posters we also invite both the academia and industry to come and show demos of their text and/or speech processing tools, apps or prototypes; this includes text or speech processing, tools for corpora, language learning, and can range from tools for specialists to end-user applications. No need to disclose all the details of how you did it (looking at you, IP-loving industry people), as long as it works!

 

To participate send us a short abstract (1-2 pages) describing the purpose and functionality of the tool/app/prototype, and which should also include the list of authors and a title. Accepted abstracts will be made publicly available on the conference webpage (unless you object), but will not be included into the conference proceedings. At the conference you will have a poster stand and a table next to it for showing your demo -- it is up to you to decide how exactly to use it.

SEPTEMBER 5 Demo abstract submission deadline

SEPTEMBER 12 Acceptance/rejection notification

The abstract does not have to be anonymous and should be sent directly to fishel@ut.ee

 

SUBMIT

YOUR PAPER

We invite you to submit extended abstracts on substantial, original, and unpublished research in the area of natural language processing and language technologies in general, including but not limited

to the following topics:

Applications:

Machine translation and multilinguality

Human-computer interaction, dialog systems, question answering

Multimodal language processing, image captioning

Speech recognition, synthesis, translation

Information extraction, natural language understanding

Handling humor/irony, slang, technical, legal, other special domains

 

Core methodology:

Machine learning and deep learning for language processing

Reinforcement/unsupervised/semi-supervised/transfer learning in NLP

Distributed representation learning and applications

Linguistic knowledge incorporation in machine learning approaches

Tagging, disambiguation, shallow/deep parsing

Ontologies, lexicons, terminology, knowledge representation

(Semi)automatic evaluation metrics and error analysis

 

Language resources:

Text/speech/multimodal data, design, evaluation

Resources and methods for lesser-resourced languages

 

Special interest: research on languages spoken in the Baltic countries

Abstracts should consist of about 1000-1500 words and will be submitted through EasyChair

and will be peer-reviewed.

Papers accepted for presentation at the conference will be included in the Proceedings of "Human Language Technologies - The Baltic Perspective", which is published by IOS Press in the series "Frontiers in Artificial Intelligence and Applications" as an Open Access book. FAIA volumes are included in some of the major A&I databases, for example, SCOPUS and Web of Science.

 

By uploading your camera-ready document you automatically agree to give the permissions to publish your paper in the Proceedings. The templates and formatting guidelines for accepted papers are the following: LaTeX / MsWord

SUBMISSION IS CLOSED

SPEAKERS

 

MARTIN VOLK

University of Zurich / Switzerland

Prof. Martin Volk is the head of the Institute of Computational Linguistics at the University of Zurich. He has successfully lead numerous projects on digital humanities, machine translation and natural language processing. His

talk will be on the topic of "Language Technology and Digital Humanities".

PETER BELL

University of Edinburgh / United Kingdom

Peter Bell is a senior research associate in automatic speech recognition at the Centre for Speech Technology Research, in the School of Informatics at the University of Edinburgh. He will talk about "Automatic speech recognition for TV broadcasts".

ALEXANDER FRASER

LMU Munich / Germany

Alexander Fraser is the Professor of Information and Language Processing at CIS, LMU Munich. He leads a research group focusing on machine learning techniques for machine translation and related NLP problems. His talk will be

on the topic of "Translation to Morphologically Rich Languages".

PROGRAM

DAY 1 / SEPTEMBER 27

WIKIPEDIA MEETS NLP

Workshop

14:00

WELCOME RECEPTION

of the 8th International Conference

“Human Language Technologies - the Baltic Perspective”

19:00

 

DAY 2 / SEPTEMBER 28

Registration

08:30

 

Opening

09:30

KEYNOTE: MARTIN VOLK

10:00

Coffee break

11:00

TALKS: MORPHOLOGY AND SEMANTICS

 

Neural Morphological Tagging for Estonian

Alexander Tkachenko and Kairit Sirts

 

Estonian Morphology in the Giella

Infrastructure

Heiki-Jaan Kaalep, Sjur Nørstebø Moshagen and Trond Trosterud

 

Topic interpretation using Wordnet

Eduard Barbu, Heili Orav and Kadri Vare

11:30

Lunch break

13:00

KEYNOTE: ALEXANDER FRASER

14:00

TALKS: MACHINE TRANSLATION

 

Impact of Corpora Quality on Neural Machine Translation

Matīss Rikters

 

Advancing Estonian Machine Translation

Matīss Rikters, Mārcis Pinnis and Roberts Rozis

15:00

Coffee break

16:00

DEMOS AND POSTERS

 

NLP-PIPE: Latvian NLP Tool Pipeline

Artūrs Znotiņš and Elita Cīrule

 

Extending Tēzaurs.lv online dictionary into

a morphological lexicon

Lauma Pretkalniņa and Pēteris Paikens

 

Czech & Slovak Corpus Resources Go

(not only) Latvian

Michal Škrabal and Vladimír Benko

Latvian Tweet Corpus and Investigation of Sentiment Analysis for Latvian

Mārcis Pinnis 

 

Deeper Error Analysis of Lithuanian

Morphological Analyzers

Loïc Boizou, Erika Rimkutė and

Jurgita Kapočiūtė-Dzikienė

 

Low-resource Translation Quality Estimation

for Estonian

Elizaveta Yankovskaya and Mark Fishel

MT Solutions for EU Council Presidencies

and Hugo.lv

Tilde

Monolingual and Cross-lingual Style Transfer

University of Tartu

Some Tools and Services at Center of Estonian Language Resources

Center of Estonian Language Resources

AlphaAI and AlphaChat Solution for Customer Support Chat Automation

AlphaBlues

TEXTA Toolkit

Texta

16:30

DINNER AND SOCIAL EVENT

at Estonian National Museum

18:00

DAY 3 / SEPTEMBER 29

Registration

08:30

KEYNOTE: PETER BELL

09:30

Coffee break

10:30

TALKS: CORPORA AND LINGUISTICS RESOURCES

 

Linguistically-motivated automatic classification of Lithuanian texts for didactic purposes

Gintarė Grigonytė, Erika Rimkutė and Jolanta Kovalevskaitė

Language Use in a Multilingual Tweet Corpus

Dmitrijs Milajevs

 

Latvian FrameNet: cross-lingual issues

Gunta Nešpore, Baiba Saulīte and Normunds Grūzītis

11:00

Lunch break

12:30

TALKS: SPEECH PROCESSING

 

Advanced Rich Transcription System

for Estonian Speech

Tanel Alumäe, Ottokar Tilk and Asadullah

 

Speech-based Identification of Children’s Gender and Age with Neural Networks

Leo Kristopher Piel and Tanel Alumäe

 

Towards a modern text-to-speech system

for Latvian

Roberts Darģis and Ilze Auziņa

 

General-purpose Lithuanian Automatic Speech Recognition System

Askars Salimbajevs and Jurgita Kapočiūtė-Dzikienė

14:00

Coffee break

16:00

DEMOS AND POSTERS

 

Self-reading Texts and Books

Meelis Mihkla, Indrek Hein and Indrek Kiissel

 

Collection of Resources and Evaluation

of Customer Support Chatbo

Andrejs Vasiļjevs and Daiga Deksne

Mirthful and Polite Laughter: Acoustic Features

Regina Sabonytė

F0 in Lithuanian: the indicator of stress, syllable accent, or intonation? 

Regina Sabonytė and Asta Kazlauskienė

The speech rhythm of the Lithuanian

and Latvian languages

Asta Kazlauskienė and Aistė Zigmantaitė


The legal aspects of using data from linguistic experiments for creating language resources 

Jane Klavan, Arvi Tavast and Aleksei Kelli

Estonian Speech Recognition Applications at TTÜ

Tallinn University of Technology

Snackable AI demo application

Snackable Inc

Customer Feedback Analysis with Feelingstream

FeelingStream

Multi-speaker Neural Speech Synthesis for Estonian

University of Tartu

16:30

The end

18:00

 

REGISTRATION

Regular registration is closed, onsite tickets will be available if you still decide to come!

Please note that you can only pay on site with a credit card via online transfer. 

 

Early bird registration until July 20th: 160 EUR / 110 EUR (student)

Regular until September 20th: 210 EUR / 145 EUR (students)

Onsite: 240 EUR

REGISTRATION IS CLOSED
 

INFORMATION

FOR PARTICIPANTS

 

SPONSORS