Speech Emotion Detection Using Deep Learning
Author : Dr. Sharayu Deote, Shreya Telang, Sidra Sheikh, Samruddhi Virkhare and Janhvi Meshram
Abstract :
The speech emotion recognition is a very exigent assignment of human computer interaction (HCI). This subject has gained so much attention in recent time and will soon achieve a high position for the requirement in coming years. In this strenuous field of speech emotion recognition many techniques have been utilized to extract emotions from signals, including many experimented speech analysis and classification techniques. In the classical way of speech emotion recognition features are extracted from the signals, pitches and frequencies of speech and then the features are selected which is known as selection module and then the emotions are recognized. This is a time consuming process so this paper gives an overview of the modern technique which is based on a simple algorithm based on feature extraction and model creation which recognizes the emotion.
These methods of signal processing and machine learning are widely used to recognize human emotions based on features extracted from facial images, video files or speech signals. Various Experiments were performed to test the accuracy of the classified features extracted from audio files. Results show that random decision forest learning of this hybrid acoustic features is highly effective for speech emotion recognition.
The objective of this research paper is to develop a system which can analyze and predict the expression of the human being. The study proves that this procedure is workable and produces valid results of around 80%.
Keywords :
Speech emotion recognition, SER, speech emotion recognition using deep learning
