Material Science plays a pivotal role in advancing various industrial sectors, including aeronautics, automation, construction, and biotechnology. The introduction of advanced and high-performance materials has significantly accelerated progress in these fields, offering new functionalities to products and enhancing economic and environmental sustainability through more efficient energy usage. High-performance materials such as high entropy alloys (HEAs) and polymer-derived ceramics (PDCs) have garnered considerable attention from both industry and researchers in recent years. However, the extensive resources required for the development of such materials, from design to synthesis and characterization, have limited the pace of discovering new high-performance materials. In response to this challenge, emerging strategies leveraging artificial intelligence (AI) for materials design show promise in accelerating the entire process, particularly in the initial design phases. The vast number of element combinations and the complexity of synthesizability conditions for HEAs have paved the way for the application of AI techniques like Generative Models. This study focuses on the deployment of a specific conditional tabular generative adversarial network (CTGAN) developed for tabular data. The generated synthetic data is based on conventional parametric design parameters used for HEAs, enabling the creation of specific datasets with conditions specified for each case, offering insights into the synthesizability of candidate materials. The study compares the calculated and generated data, employs CALPHAD simulations to assess the performance of generated samples, and verifies novel compositions in open materials databases available in the literature. This research is conducted within the framework of the European project ACHIEF, aimed at discovering novel materials for industrial processes.