• Nikhil T D
• Rahul Varadaraju
• Reuben Jacob
• Yash T Kasodariya
• Suma Swamy

This paper presents a machine learning approach for automatically detecting emotions from audio speech files. Accurately recognizing emotions expressed in speech signals has many potential applications in areas like human-computer interaction, customer service analysis, and psychological studies. Our methodology involves extracting Mel-Frequency Cepstral Coefficient (MFCC) features from the raw audio data and then training and evaluating several machine learning models to classify the emotions present. We evaluate Random Forest, Decision Tree, Naive Bayes, and Multi-Layer Perceptron (MLP) models on the Toronto Emotional Speech Set (TESS) dataset. The experimental results show that the Random Forest classifier achieves the highest accuracy of 99.643% in recognizing emotions like happiness, sadness, anger, fear, disgust and neutral state. We analyze the performance characteristics of each model and discuss potential areas for further improving emotion detection from speech audio.

Emotion detection, Audio analysis, Machine Learning, Random Forest, Decision Tree, Naive Bayes, Multi-Layer Perceptron, TESS dataset, Mel-frequency cepstral coefficients (MFCC) features.

"Emotion Detection of Audio Using Machine Learning", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 5, page no.h561-h568, May-2024, Available :http://www.jetir.org/papers/JETIR2405770.pdf

"Emotion Detection of Audio Using Machine Learning", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 5, page no. pph561-h568, May-2024, Available at : http://www.jetir.org/papers/JETIR2405770.pdf