The Gaussian mixture model (GMM) method is popular and efficient for voice conversion (VC), but it is often subject to overfitting. In this paper, the principal component regression (PCR) method is adopted for the spectral mapping between source speech and target speech, and the numbers of principal components are adjusted properly to prevent the overfitting. Then, in order to better model the nonlinear relationships between the source speech and target speech, the kernel principal component regression (KPCR) method is also proposed. Moreover, a KPCR combined with GMM method is further proposed to improve the accuracy of conversion. In addition, the discontinuity and oversmoothing problems of the traditional GMM method are also addressed. On the one hand, in order to solve the discontinuity problem, the adaptive median filter is adopted to smooth the posterior probabilities. On the other hand, the two mixture components with higher posterior probabilities for each frame are chosen for VC to reduce the oversmoothing problem. Finally, the objective and subjective experiments are carried out, and the results demonstrate that the proposed approach shows greatly better performance than the GMM method. In the objective tests, the proposed method shows lower cepstral distances and higher identification rates than the GMM method. While in the subjective tests, the proposed method obtains higher scores of preference and perceptual quality.
The single-phase voltage loss is a common fault. Once the voltage-loss failure occurs, the amount of electrical energy will not be measured, but it is to be calculated so as to protect the interest of the power supplier. Two automatic calculation methods, the power substitution and the voltage substitution, are introduced in this paper. Considering the lack of quantitative analysis of the calculation error of the voltage substitution method, the grid traversal method and MATLAB tool are applied to solve the problem. The theoretical analysis indicates that the calculation error is closely related to the voltage unbalance factor and the power factor, and the maximum calculation error is about 6% when the power system operates normally. To verify the theoretical analysis, two three-phase electrical energy metering devices have been developed, and verification tests have been carried out in both the lab and field conditions. The lab testing results are consistent with the theoretical ones, and the field testing results show that the calculation errors are generally below 0.2%, that is correct in most cases.
A novel VC (voice conversion) method based on hybrid SVR (support vector regression) and GMM (Gaussian mixture model) is presented in the paper, the mapping abilities of SVR and GMM are exploited to map the spectral features of the source speaker to those of target ones. A new strategy of F0 transformation is also presented, the F0s are modeled with spectral features in a joint GMM and predicted from the converted spectral features using the SVR method. Subjective and objective tests are carried out to evaluate the VC performance; experimental results show that the converted speech using the proposed method can obtain a better quality than that using the state-of-the-art GMM method. Meanwhile, a VC method based on non-parallel data is also proposed, the speaker-specific information is investigated using the SVR method and preliminary subjective experiments demonstrate that the proposed method is feasible when a parallel corpus is not available.
In this paper, 3 typical organic fluids were selected as working fluids for a sample slag washing water binary power plants. In this system, the working fluids obtain the thermal energy from slag washing water sources. Thus, it plays a significant role on the cycle performance to select the suitable working fluid. Energy and exergy efficiencies of 3 typical organic fluids were calculated. Dry type fluids (i.e., R227ea) showed higher energy and exergy efficiencies. Conversely, wet fluids (i.e., R143a and R290) indicated lower energy and exergy efficiencies, respectively. Słowa kluczowe