In order to help presenters to improve their oral presentation skills, we propose a support system to provide impression-related feedbacks. Our support system includes two parts: 1) a presentation impression prediction system to tell presenters how audiences feel about their presentations, and 2) a presentation slide analysis system to tell presenters whether the slide designs are good and how to improve (if necessary). For the presentation impression prediction system, we propose a multimodal neural network to predict audiences’ impressions for speech videos. For the slide analysis system, we used Convolutional Neural Network and Global Average Pooling to evaluate the design of slides. We then used Class Activation Mapping to provide visual feedback for showing which areas should be modified. We created an oral presentation support system, which effectively achieved these functions.