As part of this master's thesis, a system for distributing and evaluating musical dictate has been developed. It consists of several backend applications that serve as APIs for web applications used by teachers and students. The backend applications are built using the ASP.NET framework, while the frontend applications are implemented using the React library. Notifications and job distribution are achieved through the RabbitMq message bus with the help of the MassTransit library, and real-time communication is implemented by the SignalR library. The applications are containerized using Docker. The system enables teachers to define exercises, assign them to specific student groups, and ultimately grade them using the web application. Tasks can be automatically graded if the teacher chooses to define them as such. Students are given the opportunity to practice the tasks assigned by the teacher, thereby improving their music dictation skills and preparing for assessments. The system includes an integrated graphical interface for entering musical notation, which is available to teachers for creating exercises and to students for solving them. The system can be used by any teachers and their students. They simply need to register in the system, and they are ready to use it. If desired, it is also possible to deploy the system on their own computer or resources. The solution architecture allows for system expansion by adding additional components for automated grading.