Helmer Strik

27 Oct 2017, Training data and resources released
Below you can find the announcement.

Some URLs:

• 2nd edition: https://regulus.unige.ch/spokencallsharedtask_2ndedition/
• 1st edition: https://regulus.unige.ch/spokencallsharedtask/
• SLaTE-2017 program, incl. the special session ‘CALL shared task challenge’ : http://www.slate2017.org/program.html
• SLaTE-2017 abstract book: http://www.slate2017.org/SLaTE_abstract_book.pdf

Announcement – 2nd edition of the ‘Spoken CALL Shared Task’

Dear colleagues,

Recently, we organized the first ‘Spoken CALL Shared Task’ [https://regulus.unige.ch/spokencallsharedtask/].
Results were presented at a special session of the SLaTE-2017 workshop [http://www.slate2017.org/].

The first ‘Spoken CALL Shared Task’ was successful, people expressed interest in a follow-up, and that is why we are now organizing a 2nd edition of the ‘Spoken CALL Shared Task’ [http://tiny.cc/2nd-spoken-CALL-task ~ https://regulus.unige.ch/spokencallsharedtask_2ndedition/].
The second version of the task will offer improved resources. In particular, it will make available the highest-performing speech recognizer from the first version of the task, and about twice as much training data.

The preliminary schedule of the 2nd ‘Spoken CALL Shared Task’ is as follows:

• 27 Oct 2017, Training data and resources released
• 31 Jan 2018, Release test data
• 7 Feb 2018, Deadline for submission of entries
• 23 Mar 2018, Interspeech submission deadline
• 2-6 Sep 2018, Presentations at a special session of Interspeech-2018

If you are interested (even if you are not sure that you will [be able to] participate),
Subscribe to the Newsletter at https://regulus.unige.ch/spokencallsharedtask_2ndedition/,
and/or send an email to Johanna.Gerlach@unige.ch
There are no obligations at all, this is just to have an initial idea about who is interested.

Feel free to forward this announcement.
If you have suggestions about where we can send this announcement, please let us know.

Regards,
The ‘Spoken CALL Shared Task’ team

ISCA SIG SLaTE: Speech and Language Technology in Education

MijnST: Mijn Spraaktrainer / MyST: My Speech Trainer is een proeftuin van de RU.

A.s. vrijdag 29 en zaterdag 30 september presenteren we MijnST / MyST op Drongo in Utrecht.

Brief summary (in English):
In this project we will study how to make optimal use of an e-Learning system for training spoken English. First, the content of the NovoLanguage Player will be optimized for use at the Radboud University, by adding Academic English content, and the result is MyST: My Speech Trainer. Then international students will work with MyST. Experiences and results will be collected and studied.

Samenvatting (in het Nederlands):
Goede mondelinge communicatieve vaardigheden in het (Academisch) Engels (Academic English), incl. goede uitspraak en juiste woordcombinaties (collocations), zijn belangrijk voor steeds meer studenten en medewerkers van de Radboud Universiteit (RU). Voor het verbeteren van mondelinge vaardigheden is het noodzakelijk om intensief te oefenen, en direct feedback te krijgen op de gesproken uitingen. Met MijnST wordt dit mogelijk. Het eerste doel van deze pilot is om duidelijk te maken wat de meerwaarde is van MijnST, en inzicht te krijgen hoe MijnST het beste ingezet kan worden. Het uiteindelijk doel (na deze pilot) is een ICT-tool die gebruikt kan worden door alle studenten en medewerkers van de RU, om hun communicatieve vaardigheden (incl. uitspraak en woordenschat) te oefenen, en zo het Engels universiteitsbreed op een hoger niveau te tillen. Aanstaande studenten en medewerkers van RU uit het buitenland kunnen met MijnST ook al training krijgen in het land van herkomst. Met de editor is het mogelijk om de content te optimaliseren voor iedere doelgroep. Dit alles komt ten goede aan onderwijs en onderzoek van de gehele RU, en aan de internationalisering van de RU.

Aanvragers:

• Helmer Strik (coördinator); Taalwetenschap, CLST, NovoLanguage
• Liesbet Korebrits; Radboud In’to Languages
• Martijn Enter, Joost van Doremalen; NovoLanguage

Anderen die meegewerkt hebben:

• Medewerkers: Edith Schouten, Caro Struijke
• Studenten (voor scriptie, stage, stage-minor, etc.): Varsha Melwani, Tineke Kramer-Pijnenburg, Thu Dinh Quoc Anh, Anna Ovchinnikova, Ezgi Toprakdeviren

.

TAPAS: Training network on Automatic Processing of PAthological Speech (#766287)

Abstract
There are an increasing number of people across Europe with debilitating speech pathologies (e.g., due to stroke, Parkinson’s, etc). These groups face communication problems that can lead to social exclusion. They are now being further marginalised by a new wave of speech technology that is increasingly woven into everyday life but which is not robust to atypical speech. TAPAS is proposing a programme of pathological speech research, that aims to transform the well-being of these people.
The TAPAS work programme targets three key research problems:
(a) Detection: We will develop speech processing techniques for early detection of conditions that impact on speech production. The outcomes will be cheap and non-invasive diagnostic tools that provide early warning of the onset of progressive conditions such as Alzheimer’s and Parkinson’s.
(b) Therapy: We will use newly-emerging speech processing techniques to produce automated speech therapy tools. These tools will make therapy more accessible and more individually targeted. Better therapy can increase the chances of recovering intelligible speech after traumatic events such a stroke or oral surgery.
(c) Assisted Living: We will re-design current speech technology so that it works well for people with speech impairments and also helps in making informed clinical choices. People with speech impairments often have other co-occurring conditions making them reliant on carers. Speech-driven tools for assisted-living are a way to allow such people to live more independently.

TAPAS adopts an inter-disciplinary and multi-sectorial approach. The consortium includes clinical practitioners, academic researchers and industrial partners, with expertise spanning speech engineering, linguistics and clinical science.

List of participants:

Information on partner organisations: