Audio Sentiment Analysis Using CNN and BiLSTM

From market research to human-computer interaction, sentiment analysis in audio data is an essential activity with many applications. In this research, we worked on a unique method for sentiment analysis in audio data that uses bidirectional long short-term memory (Bi-LSTM) networks and convolutional neural networks (CNN). This research investigates how class division affects model performance. The validation in this research facilitates better comparison analysis and valuable contribution.