Task 4: Low-Resource Speech-to-Text


The goal of this task is to build a system translating Swiss German speech to Standard German text.

We’ll provide a training set of roughly 100 hours of audio with sentence-aligned text.

We encourage participants to explore suitable models, data augmentation methods and transfer learning approaches for this challenging problem.

You’ll find the competition here.


Organizers


  • Michel Plüss (FHNW, Switzerland)
  • Lukas Neukom (FHNW, Switzerland)
  • Manfred Vogel (FHNW, Switzerland)

Important Dates


  • 3 February 2020: Start of challenge (registration and submission page opens, release of data)
  • 4 May 2020: Paper submission deadline and speaker invitation
  • 15 May 2020: Paper reviews due
  • 29 May 2020: Camera-ready paper deadline
  • 23 – 25 June 2020: SwissText (presentations + discussions + prize awarding)