Improved Low-Frequency Crosstalk Cancellation in Bone Conduction Using Bone Transducers and Probe Microphone

A bone conduction (BC) sound presented on either side of the head will travel to the cochleae in both ears, and this crosstalk phenomenon is considered one factor limiting the benefits of binaural hearing with BC. Previous studies have demonstrated the feasibility of measuring the phase and level required for crosstalk cancellation at the cochlea; however, given a human subject’s inability to consistently detect low frequencies, this method is limited to cancellations at frequencies above 1000 Hz. This study describes an alternative approach for implementing a crosstalk cancellation system for low frequencies by measuring ear-canal sound pressure (ECSP). Since the ear canal is close to the cochlea, we hypothesized that altering ECSP in response to a BC stimulation may be used to achieve crosstalk cancellation at the cochlea. Our approach relies on ECSP measurements to estimate impulse responses (IRs) from bone transducers to a probe microphone in the ear canal. The IRs are then used to predict the phase and level required for crosstalk cancellation at the probe microphone’s location. To confirm the hypothesis, we measured tone reception threshold (TRT), the lowest tone level participants could detect, under two conditions: with and without crosstalk cancellation. Although crosstalk cancellation was designed to occur at the probe microphone and not at the cochlea, the TRT results showed that participants were still able to perceive cancellations at frequencies below 1000 Hz.