Hilbert spectra allow identification of instantaneous frequencies that are attributed to specific corrosion mechanisms in electrochemical noise data. The present work proposes to identify and analyze areas of interest in Hilbert spectra, which enables to obtain valuable frequency information from electrochemical noise signals. Experiments were performed on AISI304 exposed to aqueous HCl solutions at different pH values resulting in either distinct general or localized attack. Results indicate that application of the proposed transient analysis to Hilbert spectra provides a significantly improved determination of the frequency characteristics of the electrochemical noise signals compared to time-frequency data analysis without transient analysis. (C) 2013 Elsevier Ltd. All rights reserved.