The interactive tablets (pen-based or touch screen devices) offer users a new way to run commands by drawing gestures. This induces intuitive interaction but the learning of all gestures for user can be tedious when the set of commands is large. Our objectives are to allow user to draw natural handwritten gestures and to help him learn these gestures quickly. After studying some existing methods, we strongly believe that the personalization capability of gestures is really important for user. In this way, we present in this paper Customizable Gesture Menus associated with an evolving gesture recognition engine that can learn incrementally, starting from few data samples. The paper is focused on these menus that allow a simple and compact presentation of personalized gestures. Two comparative tests with 59 users are reported in this paper to evaluate this new concept.