In shaly sands and low-resistivity, low-contrast reservoirs (LRLC), Archie’s saturation model often breaks down, leading to inaccurate water saturation estimates. We introduce a new semi-empirical saturation model that estimates water saturation using compressional and shear sonic log measurements. Archie’s saturation model has been the most widely used empirical approach for estimating water saturation in oil and gas reservoirs. It relies on measurements of formation resistivity, brine salinity, and the reservoir’s electrical properties to provide accurate estimates. The model assumes that the electrical conductivity of reservoir rock is primarily due to the conductive fluids occupying the pore spaces. Alternative models have been developed to address cases involving low-salinity brines or high clay content also have limitations and may not consistently outperform Archie’s equation. Our new approach is grounded in the principle that hydrocarbon-bearing formations typically exhibit lower bulk moduli than water-saturated rocks. The model aligns with theoretical rock physics frameworks and uses velocity data and established empirical relationships between P-wave and S-wave velocities. This model is designed to complement traditional resistivity-based models in the evaluation of both conventional and unconventional reservoirs, where those models are valid. However, in LRLC reservoirs, it yields more reliable water saturation estimates than resistivity-based saturation models.