Public speaking can be challenging for individuals with Attention Deficit Hyperactivity Disorder (ADHD), due to difficulties with anxiety, focus, and thought organization. This study introduces “Confident Talk,” a speech support application that combines Artificial Intelligence and Virtual Reality to improve public speaking skills for individuals with ADHD. The Artificial Intelligence component provides real-time feedback on specific speech parameters, such as pace, tone, volume, and pauses, which are evaluated using machine learning models that analyze vocal patterns and detect irregularities in speech flow. These metrics are tailored to address common ADHD-related challenges, such as inconsistent pacing and varying voice modulation. The feedback is personalized based on each user’s progress and gradually reduces assistance as skills improve. The Virtual Reality environment, on the other hand, offers an immersive space for users to practice speaking scenarios in front of virtual audiences, enhancing their confidence and focus. By integrating these technologies, the application effectively supports skill development and caters to the unique needs of individuals with ADHD. This research contributes to the fields of educational and therapeutic technology by demonstrating the potential of Artificial Intelligence and Virtual Reality to create more effective learning experiences, with plans for future work focusing on real-world testing and user-centered design to further refine the application and maximize its impact on public speaking proficiency.