Abstract: Speech emotion recognition (SER) systems aim to recognize human emotional state during human-computer interaction. Most existing SER systems are trained based on utterance-level labels.
Some results have been hidden because they may be inaccessible to you
Show inaccessible results