Special Sessions

Phase 2 paper submission will be open till 10 Sep 2022. In Phase 2, papers are invited to the following special sessions:

  • Special Session: Data Augmentation in Speech Technologies

Abstract:

The recent advancements in research for speech technologies have opened doors towards several real-world applications. While most of these technologies perform effectively for models trained with in-domain or large amounts of training data, their performance is affected in case of mismatched test conditions or unknown settings. Data augmentation is one of the key directions that can help to increase the robustness of a model by generalizing a model’s ability to deal with mismatched/unsessn conditions. Further, it is useful to build models for applications, where sufficient training data is not available. This special session will focus on the exploration of new data augmentation methods for different speech technologies as well as the use of existing data augmentation methods for some specific speech-based applications.

Organizers:

  • Dr. Rohan Kumar Das, Fortemedia Singapore Pte. Ltd. (ecerohan@gmail.com)
  • Dr. Ville Hautamäki, National University of Singapore (villeh@nus.edu.sg)
  • Special Session: Automatic Depression Detection

  • Special Session: Audio-Visual Talking Head 

  • Special Session: Deep Noise Reduction