基于空间特征抽取与神经网络的人耳空间听觉模型

A spatial hearing model based on spatial feature extraction and artificial neural network

空间听觉中复数值的与头相关联的传递函数（ＨＲＴＦ）可用实数值的与头相关联的冲激响应（ＨＲＩＲ表示。对测量空间上归一化的ＨＲＩＲ进行Ｋａｒｈｕｎｅｎ－Ｌｏｅｖｅ展开可以提取其空间特征．用Ｖｏｎ—Ｍｉｓｅｓ函数为基函数的神经网络逼近离散的ＨＲＩＲ空间特征函数得到连续听觉空间上的双耳时域模型．模型与实际测量得到的ＨＲＩＲ有较好的一致性。

in spatial hearing, a complex valued head--related transfer function (HRTF) can be represented as a real valued head--related impulse response (HRIR). Using Karhunen-Loeve expansion, thespatial features of the normalized HRIRs on measurement space can be extracted as spatial characteristicfunctions. A neural network model based on Von--Mises function is used to approximate the discretespatial characteristic function of HRIR. As a result, a time domain binaural model is established whichwell fits the measured HRIRs.