Data Driven Beamformer Design for Binaural Headset

Mobile devices are being used in more and more adverse noise environments. This increases the requirements for designing headsets for these devices. Regardless of the limi-tations for larger battery life some designs use multiple mi-crophones to achieve appropriate noise suppression. Unfor-tunately the most common approaches for designing the beamformer do not provide good results due to the complex way that the sound generated by the mouth travels around the head to reach the area around the ear, where the headset is usually positioned. In this paper we propose a data driven approach for designing a time invariant beamformer using set of calibration files. The approach is illustrated with the evaluation of a beamformer for binaural headset. As evalua-tion criteria are used the improvements in output Signal to Noise Ratio (SNR) and objective evaluation of the percep-tual sound quality. The proposed design approach delivers 6.8 dBC improvement in SNR and 0.46 MOS points im-provement in sound quality, compared to a single micro-phone.