Authors: Chirag Sharma, Aryan Kamble, Dr. Jasbir Kaur, Assistant Professor Mr. Suraj Kanal, Assistant Professor Ms. Ifrah Kampoo
Abstract: Mental health disorders, particularly depression and anxiety, are among the leading causes of disability worldwide. Yet, early and accurate detection remains challenging due to the shortage of trained professionals, societal stigma, and the subjective nature of traditional assessments. Artificial Intelligence (AI), powered by Machine Learning (ML), Natural Language Processing (NLP), and acoustic analysis, is emerging as a promising tool for early detection. This paper presents a data-driven framework for identifying early signs of depression and anxiety through voice and text analysis. Given the extensive research in this domain, we highlight comparative studies using various AI techniques, showcasing their effectiveness in mental health diagnostics. These studies explore text-based analysis, voice-based models, multimodal fusion systems, and ethical considerations in AI-driven mental health screening. Special attention is given to real-time social media analysis and conversational voice pattern recognition, demonstrating AI’s potential in developing scalable, intelligent, and ethical preclinical screening systems. Additionally, this review presents a comparative analysis of studies employing various AI techniques, evaluating their strengths and limitations. It also highlights the accuracy of multimodal fusion models, reinforcing their superiority over unimodal approaches in mental health assessments.