Asteroseismology is used to infer the interior physics of stars. The Kepler and TESS space missions have provided a vast data set of red-giant lightcurves, which may be used for asteroseismic analysis. These data sets are expected to significantly grow with future missions such as PLATO, and efficient methods are therefore required to analyze these data rapidly. Here, we describe a machine-learning algorithm that identifies red giants from the raw oscillation spectra and captures p- and mixed-mode parameters from the red-giant power spectra. We report algorithmic inferences for large frequency separation ({Delta}{nu}), frequency at maximum amplitude ({nu}_max_), and period separation ({Delta}{Pi}) for an ensemble of stars. In addition, we have discovered ~25 new probable red giants among 151000 Kepler long-cadence stellar-oscillation spectra analyzed by this method, among which four are binary candidates that appear to possess red-giant counterparts. To validate the results of this method, we selected ~3000 Kepler stars, at various evolutionary stages ranging from subgiants to red clumps, and compare inferences of {Delta}{nu}, {Delta}{Pi}, and {nu}_max_ with estimates obtained using other techniques. The power of the machine-learning algorithm lies in its speed: It is able to accurately extract seismic parameters from 1000 spectra in ~5s on a modern computer (a single core of the Intel(r) Xeon(r) Platinum 8280 CPU).