Based on the ongoing development in the Long Term Evolution (LTE), it was brought in to satisfy user's needs and demands on the enhancement of the wireless network's Quality of Service, especially for real-time applications. Real-time applications require low delay; in order to have a better user's experience. Hence, a scheduling framework focuses on real-time applications has been proposed and developed in this study. This paper compares the performance of real-time scheduling frameworks; 1) Modified-Largest Weighted Delay First (M-LWDF), 2) Frame Level Scheduler (FLS) and Proportional Fair downlink scheduler, and 3) our proposed two-level scheduling framework (FLS-M-EXP). Tests are carried out on these scheduling methods, and the data produced are gathered and analyzed. Simulation results showed that the proposed two-level scheduling framework with FLS and Exponential Rule (EXP) has a better performance in terms of delay as compared to the existing scheduling framework, which consists of FLS and Proportional Fair method. The proposed two-level scheduling framework also performs better than the common real-time scheduler, M-LWDF.