Derio/Murcia, October 15, 2026. Xpectra, the tool developed by AI Talentum with scientific-technical backing from AZTI, represents a significant step forward in the digitalization of quality control within the food industry. This system integrates NIR spectroscopy and artificial intelligence to monitor quality and process parameters in real-time in production environments, without the need to destroy samples.
This technology reduces reliance on traditional laboratory analyses or external controls, enabling companies to anticipate quality deviations and make operational decisions with greater reliability. In its initial application to fish oil, a potential revenue increase of up to 12.5% was estimated compared to using conventional analyses exclusively.
Xpectra offers a comprehensive, turnkey solution combining scientific knowledge, artificial intelligence, specialized software, and technical support. Its differential value lies in the development of models calibrated and validated against reference methods, adapted to real production conditions and usable by non-expert personnel.
Idoia Olabarrieta, senior researcher at AZTI, explains that "NIR spectroscopy allows obtaining information about product quality quickly and without destroying the sample. The main scientific challenge is to develop robust models, calibrated against reference methods and adapted to real production conditions." In this first application, batches of fish oil produced throughout the year were analyzed, correlating their spectral fingerprint with laboratory results.
This system also enables non-destructive testing, which is a significant advantage for high-value products or frequent production line controls, approaching the inspection of the entire production.
Antonio Vicente Contreras, CEO of AI Talentum, notes that "the results obtained in fish oil quality control show that the combination of NIR spectroscopy, artificial intelligence, and a platform accessible to plant personnel can meet real industry needs."
Xpectra offers the food value chain a path to advance in digitalization, improve analysis traceability, reduce reliance on external laboratories, and accelerate decision-making. In the fish oil pilot, improved quality-based production classification and management indicates a potential revenue increase of up to 12.5%.
