Voice-Driven Emotion Recognition: Integrating Speaker Diarization for Enhanced Analysis

In today's digital age, where online communication has become indispensable, especially in light of the COVID-19 pandemic, ensuring effective communication is paramount. Traditional emotion recognition systems, predominantly reliant on facial expressions, face inherent challenges in online environments where participants often opt for camera-disabled interactions. Addressing this gap, our research introduces a novel emotion recognition approach rooted in voice data. By integrating speaker diarization models, we enable intricate emotion analysis from voice data and juxtapose these findings with subjective evaluations from participant questionnaires. Preliminary results suggest that voice-based emotion recognition can potentially revolutionize the quality of online communication, laying the groundwork for future adaptive feedback systems.