In this paper an approach is presented where acoustics and multivariate data analysis were applied to pharmaceutical inhalation device characterization. A feasibility study was performed where a set-up of two miniaturized condenser microphones and an analog-to-digital (A/D) converter was used to monitor the sound of inhalation devices being analysed in either a manual or in an automated delivered dose analysis station. A rig was used to incorporate the Acoustic Emission (AE) system, offering flexibility regarding the placement of the microphones. The raw sound data was Fourier transformed to frequency spectra, which were fed into the multivariate analysis software. The main aim of the study was to assess whether AE could be used to derive device performance information, but also act as a system suitability test of the analytical system. It was shown that this is a powerful technique capable of distinguishing between filled and empty cavities, for example, and between powder mixtures with different properties in the inhalation device, giving an indication on the type and amount of formulation that is used. Valuable metadata (e.g. detecting small differences in pressure drop) related to the quality of the analytical process could also be acquired. AE can be used to detect differences and provide valuable data on several levels, e.g. between two batches of inhalers or between two inhalers from the same batch exhibiting minor differences in physical properties that may affect the inspiratory sequence. Early exploration results on AE as a design aid in the development of a new prototype dry powder inhaler (DPI), is presented here. Special attention is paid to what possibilities various chemometric approaches can offer in terms of data handling of the information rich acoustic signal.