A Review And Analysis of Emotion Based Harmful Speech Detection Using Machine And Deep Learning And Future Direction

Nowadays, harmful speech is among the most difficult issues deal by internet. Hand-reporting comments and posts suspected of harmful speech on the internet is a frequent method of dealing with web suspects of harmful speech. This has many drawbacks. Because it necessitates the involvement of a human, this process takes a long time. Corpora of speech signals, a wide range of speech attributes, and systems for recognizing emotions are all included in this collection, which examines the most available research on speech emotion recognition. This paper presents the review of some extraordinary works done in this field. Furthermore, a short analysis is over the machine learning and deep learning algorithms. The state vector machine algorithm found to be very useful in the speech emotion recognition technique. The novel analysisand survey found to be very useful in for the upgraded researches. The paper also touched on the necessity of using a variety of classification models. If a future study is to be done on emotion recognition in general or on hazardous speech detection in particular, certain difficulties have been noted.