Teleoperation is a crucial aspect of robotics, which enables humans to control objects from a remote location by simulating the conditions of the object’s actual location. In bilateral teleoperation, force feedback is necessary for the master operator to ensure accurate work, but the transmission delay in communication can cause instability and difficulty. Despite the development of various methods to alleviate the effects of time delay in signal transmission, it remains a significant challenge for ensuring stability and transparency in bilateral teleoperation systems. One potential solution to address this challenge is the utilization of neural networks (NN). These networks are adaptable to system uncertainties and disturbances as they are model-free and can be utilized for predicting teleoperation signals. This study focuses mainly on developing a deep neural network-based control framework for the considered bilateral teleoperation system, with the presence of TOP delay and other disturbances. Long short-term memory (LSTM), a specialized variant of recurrent neural network (RNN), was driven due to its capacity to handle more intricate scenarios in contrast to traditional CNN strategies, hence it was adopted for this application. The system is modeled using Lyapunov functions and an LSTM-based controller is developed to counter the effects of TOP delay and random perturbation and measurement noises in the system. The model was developed in Python, with the use of the Keras API module of Tensorflow to use the pre-trained LSTM model.