Towards Robust Deep Neural Networks for Affect and Depression Recognition from Speech

Intelligent monitoring systems and affective computing applications have emerged in recent years to enhance healthcare. Examples of these include assessment states such as Major Depressive Disorder (MDD). MDD describes the constant expression certain emotions: negative emotions (low Valence) lack interest Arousal). High-performing intelligent would diagnosis its early stages. In this paper, we present a new deep neural network architecture, called EmoAudioNet, for emotion depression recognition from speech. Deep EmoAudioNet learns time-frequency representation audio signal visual spectrum frequencies. Our model shows very promising results predicting affect depression. It works similarly or outperforms state-of-the-art methods according several evaluation metrics on RECOLA DAIC-WOZ datasets arousal, valence, Code is publicly available GitHub: https://github.com/AliceOTHMANI/EmoAudioNet.