The recording of Transient Evoked Otoacoustic Emissions (TEOAEs) has become one of the most important clinical tests of the functionality of the inner ear, but up to now, TEOAEs analysis still relies mostly on a visual inspection of the signal in the time domain. Because of the non-stationary and non-linear nature of the TEOAEs, proper description of the TEOAEs requires simultaneous localization of signal’s structures in time and frequency domains. Existing methods of time-frequency representation of TEOAEs suffer from drawbacks such as limited time or frequency resolution and artefacts that make them impractical for practical use. It will be presented a new method for TEOAEs analysis using recently proposed Hilbert- Huang transform (HHT). The method consists of the empirical mode decomposition (EMD), which is then decomposed into a set of intrinsic mode functions (IMFs). The Hilbert transforms can then be applied to the IMFs. The local energies and the instantaneous frequencies derived from the IMFs through Hilbert transforms can be used to construct a full energy-frequency-time distribution of the data. The results obtained using the proposed method are compared to other time-frequency representations (short time Fourier transform, smoothed Wigner-Ville distribution and wavelet transform). The results show that the HHT method is superior to other time-frequency representations and gives the most precise definition of particular events in time-frequency space. The HHT is expected to be useful tool for analysing TEOAEs in a clinical practice.