• Mohamad Emad Bitar
• Dr. V. Sujatha

This paper's main goal is to provide a speech emotion recognition framework. Many aspects of the speech can be extracted to identify emotions. The removal of sentiments from waves in SERs has been accomplished using a variety of tried-and-true discourse evaluation and classification techniques. In this study, a number of challenges connected to recent work on speech-based emotion identification are described. The selection of a database, identifying numerous speech-related variables, and making an appropriate classification model choice are the main problems in emotion recognition. We have conducted literature research on the many traits that are utilized to identify emotions in human speech. The importance of several classification models has been highlighted in addition to some recent study work reviews. . Feature Extraction: A little amount of information from the speech signal is removed for analysis without altering the speech's characteristics [4]. The feature parameters known as Mel frequency cepstral coefficients are commonly utilized in voice recognition [5]. MFCC employs the Mel scale [6] based on the hearing organ's response. In this study, various features are retrieved from the speech signals, and analysis are then done on those features [7]. Max-pooling, an activation function, and many convolutional layers are required for feature extraction. The accuracy of a device's ability to recognize speech emotions is improved by using the various feature extraction methods. The focus of the study is on the pre-processing of the acquired audio samples, where noise is removed using filters from speech samples.

Speech Emotion Recognition; SVM; Classification; CNN; Mel frequency cepstral coefficients

"speech emotion recognition : investigation survey", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.9, Issue 10, page no.d293-d296, October-2022, Available :http://www.jetir.org/papers/JETIR2210350.pdf

"speech emotion recognition : investigation survey", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.9, Issue 10, page no. ppd293-d296, October-2022, Available at : http://www.jetir.org/papers/JETIR2210350.pdf