In this paper, we present an enhancement method for degraded historical handwritten documents. Document enhancement is focused on improving the text quality in document images, and our approach focuses on both denoising and improving the text quality at the same time. We use a generative adversarial network (GAN) model and aim to holistically enhance and denoise the input image and generate a high-quality output image. We tested our model on datasets of different styles and languages and obtained excellent results. In addition, we compare our model with various other approaches and show that our model outperforms them. Throughout different experiments, we show that our model has strong generalization and can be used on datasets of different languages and styles, and can handle degraded historical documents.