We propose a method for constructing a virtual sound source (VSS) whose perceived distance and direction can be controlled, using multiple pairs of parametric array loudspeakers (PALs) and electro-dynamic loudspeakers (EDLs). This method is based on the direct-to-reverberant ratio (DRR) of room impulse responses (RIRs), which serves as a key auditory cue for distance perception. PALs exhibit super-directivity by utilizing arrays of ultrasonic transducers. As a result, PALs produce RIRs with high DRRs, characterized by large-amplitude direct sounds and low-amplitude reverberation. This makes the VSS perceived closer to the listener. In contrast, EDLs produce lower DRRs, causing the VSS to be perceived as farther away. The proposed method synthesizes RIRs corresponding to arbitrary perceived distances by reproducing a target DRR at the listener’s position through a weighted combination of the RIRs from PALs and EDLs. Additionally, the direction of the VSS is controlled using vector-based amplitude panning. To improve perceptual accuracy in the near field, where DRR deviations are more pronounced, a correction function based on regression analysis of measured RIRs is introduced. We demonstrated the effectiveness of the proposed method through objective DRR measurements in various reverberant environments and subjective experiments on distance and directional perception.
Yamasaki et al. (Wed,) studied this question.