CLIPDraw: Exploring Text-to-Drawing Synthesis through Language-Image Encoders