The frequency domain power spectra of acoustic emission (AE) signals from different metal-acid reaction processes such as 6111 Al-alloy-hydrochloric acid (HCl) and 7070 Al-alloy-HCl for evolving hydrogen gases were obtained by fast Fourier transform (FFT) program and used for chemical analysis of different metal materials. Averaged power spectra from these processes and their corresponding characteristics were extracted. The characteristic AE frequency signals could be used for chemical pattern recognition of different metal materials like 6111 and 7050 aluminum alloys from the metal-acid reaction processes, that the principal component analysis (PCA) with appropriate frequency selection procedure gave a satisfactory classification with a correct rate of 78.1%. The back-propagation (BP) algorithm of artificial neural network (ANN) could give better recognition of AE signals for 6111 and 7050 alloys with a correct rate of 100%. Moreover, the AE energetic parameters are linearly correlated with the pH value of the acidic buffer solution, which opens a new possibility for quantitatively analytical application of AE signals on metal materials.