Gravitational-wave (GW) interferometers encounter significant challenges from transient noise artefacts, known as glitches. These glitches impair the sensitivity and data quality, complicating the detection of GW signals. To enhance detection capabilities, better modelling and inclusion of glitches in large-scale studies are essential. In this chapter, we explore the application of Generative Adversarial Networks (GANs), a cutting-edge deep learning algorithm, to learn the distribution of blip glitches and generate artificial populations. By reconstructing glitches in the time domain, we provide a smooth input for the GAN, enabling the creation of approximately \(10^3\) glitches from Hanford and Livingston detectors in less than one second. The performance and quality of the generated glitches are assessed using several metrics. We also introduce gengli, a user-friendly open-source software package that includes practical examples of the trained network’s usage. Furthermore, we demonstrate a practical application to improve the understanding of Gravity Spy’s performance, one of the current state-of-the-art glitch classifiers. Future work aims to extend this methodology to other glitch classes, ultimately creating an open-source interface for mock data generation. This will enhance stress testing of search pipelines and increase confidence in GW signal detection.