Bacterial infections are the leading cause of morbidity and mortality in the world, particularly due to a delay in treatment and misidentification of the bacterial species causing the infection. Therefore, rapid and accurate identification of these pathogens has been of prime importance. The conventional diagnostic techniques include microbiological, biochemical, and genetic analyses, which are time-consuming, require large sample volumes, expensive equipment, reagents, and trained personnel. In response, we have now developed a paper-based ratiometric fluorescent sensor array. Environment-sensitive fluorescent dyes (3-hydroxyflavone derivatives) pre-adsorbed on paper microzone plates fabricated using photolithography, upon interaction with bacterial cell envelopes, generate unique fluorescence response patterns. The stability and reproducibility of the sensor array response were thoroughly investigated, and the analysis procedure was refined for optimal performance. Using neural networks for response pattern analysis, the sensor was able to identify 16 bacterial species and recognize their Gram status with an accuracy rate greater than 90%. The paper-based sensor was stable for up to 6 months after fabrication and required 30 times lower dye and sample volumes as compared to the analogous solution-based sensor. Therefore, this approach opens avenues to a state-of-the-art diagnostic tool that can be potentially translated into clinical applications in low-resource environments.