Abstract Background/Introduction The healthcare sector faces an unprecedented challenge with the surging demand in both public and private settings, pushing them towards the brink of overload. This scenario underscores a critical knowledge gap in leveraging technology to alleviate the burden on healthcare services. Our study introduces an innovative solution through the development of avatars generated with artificial intelligence (AI) technologies, that replicate the likeness and expertise of healthcare professionals. These AI-generated avatars aim to bridge the gap in patient consultations, offering a scalable and efficient alternative to traditional care models. Purpose The primary objective of our study was to design and implement a virtual assistant capable of conducting patient consultations with the same reliability and personal touch as a trusted human medical professional. This was carried out through the integration of diverse AI technologies including generative models, large language models (LLMs), speech-to-text, and text-to-speech algorithms. Methods Our methodology encompassed the creation of three-components, the visual and auditory representation of the healthcare professional (avatar), the capability to process and generate speech (brain functions), and a database of information sourced from medical repositories as well as expert inputs (compiled knowledge). The development process involved capturing and generating the likeness and voice of a medical expert, training LLMs with the compiled knowledge, and engineering the system to function in real-time. This multifaceted approach ensured the AI-generated avatars were not only personalised but also intelligent and interactive. Results The successful execution of each component confirmed its feasibility, with the AI-generated avatars resembling their human counterparts in appearance and voice, effectively processing spoken inputs, and providing medical advice based on the compiled knowledge. The primary challenge identified was the integration of these components to operate seamlessly in real-time, indicating the need for further refinement in synchronisation and response speed. Conclusion This study has laid a foundational step towards revolutionising patient care through AI technologies. Our results demonstrate the potential of AI-generated avatars to alleviate workloads by providing personalised and efficient patient consultation services. Future work will focus on enhancing the real-time integration of system components, improving the scalability of the solution, and further personalising the patient experience. This direction not only intends to expand the capabilities of healthcare providers but also opens new avenues for patient engagement and care delivery.