A new customization approach based on support vector regression （SVR） is proposed to obtain individual headrelated impulse response （HRIR） without complex measurement and special equipment. Principal component ana...A new customization approach based on support vector regression （SVR） is proposed to obtain individual headrelated impulse response （HRIR） without complex measurement and special equipment. Principal component analysis （PCA） is first applied to obtain a few principal components and corresponding weight vectors correlated with individual anthropometric parameters. Then the weight vectors act as output of the nonlinear regression model. Some measured anthropometric parameters are selected as input of the model according to the correlation coefficients between the parameters and the weight vectors. After the regression model is learned from the training data, the individual HRIR can be predicted based on the measured anthropometric parameters. Compared with a back-propagation neural network （BPNN） for nonlinear regression, better generalization and prediction performance for small training samples can be obtained using the proposed PCA-SVR algorithm.展开更多
For Virtual Reality(VR) to be truly immersive, it needs convincing sound to match. Due to the diversity of individual's anthropometric measurements, the individualized customization technology is needed to get con...For Virtual Reality(VR) to be truly immersive, it needs convincing sound to match. Due to the diversity of individual's anthropometric measurements, the individualized customization technology is needed to get convincing sound. In this paper, we proposed a simple and effective method for modeling relationships between anthropometric measurements and Head-related Impulse Response(HRIR). Considering the relationship between anthropometric measurements and different HRIR parts is complicated, we divided the HRIRs into small segments and carried out regression analysis between anthropometric measurements and each segment to establish relationship model. The results of objective simulation and subjective test indicate that the model can generate individualize HRIRs from a series of anthropometric measurements. With the individualized HRIRs, we can get more accurate acoustic localization sense than using non-individualized HRIRs.展开更多
In spatial hearing, complex valued head-related transfer function (HRTF) can be represented as a real valued head-related impulse response (HRIR). By using Karhunen-Loeve expansion, the spatial features of the normali...In spatial hearing, complex valued head-related transfer function (HRTF) can be represented as a real valued head-related impulse response (HRIR). By using Karhunen-Loeve expansion, the spatial features of the normalized HRIRs on measurement space can be extracted as spatial character functions. A neural network model based on Von-Mises function is used to approximate the discrete spatial character function of HRIR. As a result, a time-domain binaural model is established and it fits the measured HRIRs well.展开更多
A head-related transfer function （HRTF） model for fast and real-time synthesizing multiple virtual sound sources is proposed. A head-related impulse response （HRIR, time- domain version of HRTF） is first decompose...A head-related transfer function （HRTF） model for fast and real-time synthesizing multiple virtual sound sources is proposed. A head-related impulse response （HRIR, time- domain version of HRTF） is first decomposed by a two-level wavelet packet and then represented by a model composed of subband filters and reconstruction filters. The coefficients of the subband filters are the zero interpolation of the wavelet coefficients of the HRIR. The coefficients of the reconstruction filters can be calculated from the wavelet function. The model is simplified by applying a threshold method to reduce the wavelet coefficients. The calculated results indicate that for a model with 30 wavelet coefficients, the error of reconstructed HRIR is about 1%. And the result of a psychoacoustic test shows that a model with 35 wavelet coefficients is perceptually indistinguishable from the original HRIR. When multiple virtual sound sources are synthesized simultaneously, the computational cost of the proposed model is much less than the traditional HRTF filters.展开更多
基金Project supported by the Shanghai Natural Science Foundation (Grant No.08ZR1408300)the Shanghai Leading Academic Discipline Project (Grant No.S30108)
文摘A new customization approach based on support vector regression （SVR） is proposed to obtain individual headrelated impulse response （HRIR） without complex measurement and special equipment. Principal component analysis （PCA） is first applied to obtain a few principal components and corresponding weight vectors correlated with individual anthropometric parameters. Then the weight vectors act as output of the nonlinear regression model. Some measured anthropometric parameters are selected as input of the model according to the correlation coefficients between the parameters and the weight vectors. After the regression model is learned from the training data, the individual HRIR can be predicted based on the measured anthropometric parameters. Compared with a back-propagation neural network （BPNN） for nonlinear regression, better generalization and prediction performance for small training samples can be obtained using the proposed PCA-SVR algorithm.
基金supported by the National Key R&D Program of China(No.2017YFB1002803)the National Nature Science Foundation of China(No.61671335,No.U1736206,No.61662010)the Hubei Province Technological Innovation Major Project(No.2016AAA015)
文摘For Virtual Reality(VR) to be truly immersive, it needs convincing sound to match. Due to the diversity of individual's anthropometric measurements, the individualized customization technology is needed to get convincing sound. In this paper, we proposed a simple and effective method for modeling relationships between anthropometric measurements and Head-related Impulse Response(HRIR). Considering the relationship between anthropometric measurements and different HRIR parts is complicated, we divided the HRIRs into small segments and carried out regression analysis between anthropometric measurements and each segment to establish relationship model. The results of objective simulation and subjective test indicate that the model can generate individualize HRIRs from a series of anthropometric measurements. With the individualized HRIRs, we can get more accurate acoustic localization sense than using non-individualized HRIRs.
基金The work is supported by National Natural Science Foundation of China !(Grant No. 69871009).
文摘In spatial hearing, complex valued head-related transfer function (HRTF) can be represented as a real valued head-related impulse response (HRIR). By using Karhunen-Loeve expansion, the spatial features of the normalized HRIRs on measurement space can be extracted as spatial character functions. A neural network model based on Von-Mises function is used to approximate the discrete spatial character function of HRIR. As a result, a time-domain binaural model is established and it fits the measured HRIRs well.
基金supported by the National Nature Science Fund of China(50938003,10774049)State Key Lab of Subtropical Building Science,South China University of Technology
文摘A head-related transfer function （HRTF） model for fast and real-time synthesizing multiple virtual sound sources is proposed. A head-related impulse response （HRIR, time- domain version of HRTF） is first decomposed by a two-level wavelet packet and then represented by a model composed of subband filters and reconstruction filters. The coefficients of the subband filters are the zero interpolation of the wavelet coefficients of the HRIR. The coefficients of the reconstruction filters can be calculated from the wavelet function. The model is simplified by applying a threshold method to reduce the wavelet coefficients. The calculated results indicate that for a model with 30 wavelet coefficients, the error of reconstructed HRIR is about 1%. And the result of a psychoacoustic test shows that a model with 35 wavelet coefficients is perceptually indistinguishable from the original HRIR. When multiple virtual sound sources are synthesized simultaneously, the computational cost of the proposed model is much less than the traditional HRTF filters.
