A Speech Emotion Recognition System Based on FPGA and ANN Network

All human beings experience a variety of emotions on a daily basis, including but not limited to joy, sadness, fear and anger. These emotions are often expressed through speech, and when individuals speak, the emotional state is embedded in the language. This paper proposes a speech emotion recognition system that is based on FPGA (field-programmable gate array) and PNN (Probabilistic Neural Network). The system employs a multifaceted approach to recognise emotions through the analysis of speech signals. This process involves several stages, including signal capture, pre-processing, feature extraction, and classification. Two artificial neural networks, PNN and LVQ, were investigated, with PNN used to process and analyse speech data, ensuring efficient data processing and real-time emotion recognition using FPGA. Finally, the system uploads the data to the cloud for deep processing and continuously optimises the emotion recognition model through a user feedback mechanism to improve personalisation and accuracy.