基于人工混响的声源空间距离恢复模型

Sound source distance reproduction model based on artificial reverberation

视听系统中3D音视频的不匹配严重影响用户体验,尤其是三维声场中对感知距离的重建技术亟待突破,为此,根据3D视频场景中已知的声源空间距离信息,结合人耳距离感知机理,采用人工混响方法控制音频信号的直达声和混响声的能量比,提出一种基于人工混响的声源空间距离恢复模型,对声源距离进行恢复。实验结果表明,该方法可以实现声源距离的恢复,恢复效果与采用最新的多距离HRTF库时相当,存储需求更低,实现更简单,实用性更强。

The mismatch of the 3D video and 3D audio in audiovisual system seriously affects the user experience, especially the reproduction of perceptual distance is urgent to overcome. For this reason, a model that according to the existing sound source distance information from the 3D video, and combined with the mechanism of human distance perception while using the artificial reverberation controlling the audio signal of direct to reverberant energy ratio to reproduce the sound source distance, was presented. Experimental results show that the proposed method can reproduce sound source distance and it nearly has the same effect with the method using the latest multi-distance HRTF database. The proposed method has lower storage requirements and it is more simple and practical.