The objective of this study is to estimate the concentration of selected microbial volatile organic compounds (mVOCs), namely ethylbenzene, cumene, and limonene, using multi-sensor gas arrays combined with statistical regression and machine learning methods. Two sensor array configurations were employed: metal oxide semiconductor (MOS) sensors and electrochemical (EC) sensors. Measurements were conducted under both field conditions, in buildings with varying degrees of microbial contamination, and laboratory conditions using fungal cultures. The results demonstrate that sensor arrays, combined with models such as Support Vector Machines (SVM) and Multi-Layer Perceptron (MLP), can effectively estimate concentrations of mVOCs with different physicochemical properties and concentration ranges. The findings confirm the potential of low-cost sensor systems as an alternative to conventional chromatographic techniques for indoor air quality monitoring.