In recent years, the concept of Mobility-as-a-Service has significantly impacted the transportation sector by integrating diverse modes of transport into a user-friendly experience. The advancement of human mobility patterns has been facilitated by the utilisation of mobile sensing technologies, but this progress has also raised concerns regarding privacy and the management of data. This study suggests increasing the applicability of human mobility data by generating synthetic data with deep learning models trained on the existing dataset. Our approach aims to enhance the practicality of human mobility data. The produced synthetic data encompasses real-world dynamics and give possibility to develop and evaluate the algorithms for personalised travel recommendations, while safeguarding sensitive information. Exploring this domain has the potential to bring about a paradigm shift in the field of mobility solutions that prioritise privacy, efficiency, and user satisfaction, ultimately leading to the development of a sustainable urban mobility framework.