Joint Actions for Swiss German
This workshop is the third meeting of a growing interest group gathered around development and use of resources for automatic processing of Swiss German. The first workshop brought together numerous interested parties providing an overview of the existing resources and potential developments. The outcome of the second workshop was a plan for developing concrete joint projects and an initial coordination of the joint activities. These activities are directed towards developing a roadmap for Swiss German NLP, where two topics were identified as high priorities: speech-to-text and writing normalisation. This resulted, among other things, in a proposal for a Bridge Discovery project for speech-to-text (submission in April 2019).
The goal of this workshop is to assess any other progress made so far, and to determine the next steps. The workshop will consist of three parts:
– The first part will be a short briefing session covering the activities undertaken since the last meeting.
– The second part will be dedicated to elaborating new ideas for the next steps. For this, we will split into smaller working groups:
1. Gathering and managing the data needed for training speech-to-text models. Here, the planned outcomes are a) specifications of a common data format; b) recording standards; and c) a data acquisition plan.
2. Writing normalization.
3. Roadmap for Swiss German NLP technologies and research.
– The third (short) part will deal with organisational issues and planning.
The workshop is open to anybody interested in NLP technologies for Swiss German, both from academia and industry.
Organizers
Tanja Samardžić, Mark Cieliebak, Jan Deriu