An investigation of machine learning techniques in speech emotion recognition

Anu Saini, Amit Ramesh Khaparde, Sunita Kumari, Salim Shamsher, Jeevanandam Joteeswaran, Seifedine Kadry


The natural languages are medium of communication from the inception of civilization. As the technology improves, the text messages, voice messages and videos are the addons in medium of communication. In long distance communication, the analysis of expression is modern area of research. The parameters of assessment are subjective hence the emotion recognition is challenging task. This article furnishes the investigation of various machine learning techniques and novel methods for speech emotion recognition (SER) to determine the feeling/sentiments in a speech. Here, we investigate the three machine learning methods named multinominal Naive Bayes (MNB), logistic regression (LR), and linear support vector machine (LSVM). Further, these techniques are incorporated with the proposed method. The performance of these machine learning techniques is investigated on two different datasets.  The datasets consist of voice and text data samples. The prosed method is trained and tested on these datasets. As per the experimentation, it has been observed that the LSVM has outperformed the other two machine learning techniques.


Deep learning; Emotions; Machine learning; Sentiment; Speech emotion recognition

Full Text:




  • There are currently no refbacks.

Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

shopify stats IJEECS visitor statistics