Deep learning of children's attitude

The optimal DeepDag executes the classification tasks in the following order:

(i) enthusiastic vs. apathetic (CNN+logMel)

(ii) enthusiastic vs. uncertain (CNN+logMel)

(iii) enthusiastic vs. confident (CRNN+logMel)

(iv) apathetic vs. confident (CRNN+logMel)

(v) apathetic vs. uncertain (CNN+logMel)

(vi) confident vs. uncertain (CNN+logMel)

Stavros Ntalampiras, “Deep Learning of Attitude in Children’s Emotional Speech”, IEEE International Conference on Computational Intelligence and Virtual Environments for Measurement Systems and Applications, 22-24 June 2020, Tunis, Tunisia. Winner of best regular paper award