At VRAIN, research in machine learning and deep learning encompasses visual artificial intelligence, deep neural networks, and advanced perception systems. Work in this area spans computer vision for image and video analysis, multimodal interaction systems that combine visual, audio, and textual information, and immersive technologies including virtual reality (VR), augmented reality (AR), and extended reality (XR).

Deep neural network architectures — including acoustic models, language models, and neural vocoders — support speech recognition and synthesis across nine languages: Spanish, Catalan/Valencian, English, German, Portuguese, Italian, Dutch, French, and Slovenian. Research in this area also develops explainable AI systems for model interpretation, data mining and analysis tools, and recommendation systems. AI evaluation frameworks are used to assess model quality and identify failure modes under real-world operating conditions, with applications in healthcare decision support, industrial quality control, and immersive training simulations.