Optimized Sentiment Analysis in Tagalog Speech Using PCA and BRNN on Prosodic Suprasegmental and MFCC Features

Sentiment analysis in Tagalog Speech is a field that has seen growing interest and development but still faces several challenges and opportunities for enhancement. This study focuses on the development of a Tagalog speech sentiment analysis model utilizing the hybrid prosodic suprasegmental features and Mel Frequency Cepstral Coefficients (MFCCs) optimized by Principal Component Analysis (PCA) and Bidirectional Recurrent Neural Network (BRNN) architecture. Prosodic suprasegmental and MFCC features which capture the spectral content and emotional tone of speech are effectively integrated. Another significant improvement is the use of PCA for dimensionality reduction, which enhanced the model's performance by reducing overfitting. The model achieved a higher accuracy rate of 90.91% compared to the 82% of existing development after being trained to classify sentiments into positive, negative, and neutral categories. The findings of the study show the potential of sophisticated machine learning methods for speech sentiment analysis, especially for Tagalog language where research is relatively limited.

I. Introduction

Speech sentiment is progressing and emerging in different fields of research. It is the computational examination of individuals' views [1], feelings, evaluations, and attitudes toward something[2]. Speech sentiment analysis has garnered a lot of attention lately because of its many uses in a variety of fields, including market research, social media monitoring, student and customer feedback analysis, and more, to find out what people think and desire [3]. It is the process of categorizing the emotional tones that speakers convey through their spoken words [4] important details regarding the views, convictions, and feelings of people speaking can be gleaned from this procedure. Speech or audio sentiment analysis is becoming more and more popular as speech data becomes more readily available and people communicate with one another via speech.