Solid-state optoelectronic sensor, based on a novel concept of integrating steady-state (SS) and time-resolved measurements (TRM) in a miniature multi-spectral filterless design, capable of detection and quantification of several important analytes, is presented. Light-emitting diodes (LED) and wavelength-selective photodiodes (PD) covering the entire range of wavelengths from UV to near-IR are used for excitation and detection, respectively. An artificial neural network (ANN) trained by signature chemical library data collected by the sensor, is used for data classification. This library is periodically updated to include additional signatures and subsequent ANN training. The experiments using experimental SS data combined with typical TRM data resulted in a classification accuracy of more than 96% for 38 categories. For analytes with similar SS characteristics but distinct TRM features, addition of lifetime data in ANN classification resulted in improved classification accuracy by about 10% over SS data alone. Major progresses in the field of photodiode technology have led to realization of a complete solid-state TRM system.