Sign language detection is essential for improving accessibility, communication, and inclusion for individuals who are deaf or hard of hearing. This technology enables smooth communication across various environments, including schools, workplaces, healthcare settings, and everyday interactions. By accurately interpreting sign language gestures, it bridges communication barriers, providing equal access to information and services. In education, sign language detection supports language learning and academic success for deaf students, while in healthcare, it ensures that medical information is conveyed accurately, enhancing care quality. In the workplace, it fosters inclusivity by enabling communication between deaf employees and their colleagues, promoting equal opportunities and professional growth. Additionally, during emergencies, it can save lives by enabling swift communication with emergency responders. This study presents a robust sign language detection system based on convolutional neural networks (CNNs), trained on a dataset of 2,515 images representing 36 distinct ASL gestures. The system employs a comprehensive preprocessing pipeline, with images standardized, resized, and split for training and validation, resulting in an accuracy rate of 85.87% across ASL letters and numerals. Real-time testing, facilitated by OpenCV’s video capture, demonstrates the model’s effectiveness in live gesture recognition with minimal delay. Despite minor errors in misclassification and occlusion, the system shows strong potential to support inclusive communication across settings and to further advance sign language recognition technology.