Glottal inverse filtering is often used to recover the glottal flow waveform from the produced voice outcomes. While there have been many studies aiming to evaluate its validity, few studies investigated its validity in voice conditions with strong source-filter interaction. The goal of this study is to evaluate the performance of glottal inverse filtering in conditions of epilaryngeal tube narrowing and/or reduced lip opening, both of which are expected to enhance source-filter interaction. Computational simulation of voice production was performed using a three-dimensional vocal fold model coupled to a vocal tract with varying degrees of constriction at the epilarynx and lips. Different algorithms of inverse filtering were then used to estimate the glottal airflow waveform from either the produced voice or airflow at the lips, and compared to the true glottal airflow from the simulation. The estimation errors in the glottal flow waveform and selected glottal flow-based measures will be evaluated as a function of the epilaryngeal and lip openings.