Objective evaluation of static beamforming on the quality of speech in noise

Beamforming is commonly used in devices such as in hearing instruments (HI) with a view to reducing noise. It consists in focusing on audio signals arising from a prescribed steering direction while suppressing those coming from other directions. The efficiency of beamformers on speech intelligibility has been demonstrated in various studies. Nevertheless, little is known about the effect on speech quality. This paper aims to assess the impact of static beamforming on the signal statistics and on speech quality in the context of hearing aids using four objective measures. Three of them are associated to temporal fine structure (TFS), spectral content evolution and envelope information, respectively. The well-known Hearing-Aid Speech Quality Index (HASQI) is used to assess speech distortion additionally. These assessments have been performed in various acoustic conditions in order to evaluate the impact of the room, the signal-to-noise ratio (SNR), and the direction of arrival of the speaker’s voice. The measures indicate that the resort to beamforming tends to improve listening quality in addition to the speech intelligibility enhancement it provides.