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.

Tris(8-hydroxyquinoline)aluminium with poly(N-vinylcarbazole) (Alq ₃:PVK) or polystyrene sulfonate (Alq ₃:PSS) were deposited by spin-coating on glass and silicon substrates. SEM measurements show that relatively smooth thin films were obtained. Fourier transform infrared measurements were performed to confirm the composition of the samples. The optical properties of thin films containing Alq ₃:PVK and Alq ₃:PSS were characterised using absorption spectroscopy and spectroscopic ellipsometry. It was found that the absorption spectrum of Alq ₃:PVK is characterised by four bands, while for Alq ₃:PSS only three bands are visible. The photoluminescence of the studied thin layers shows a peak with a maximum at about 500 nm. Additionally, cyclic voltammetry of Alq ₃ is also presented. Theoretical density functional theory calculations provide the insight into the interaction and nature of Alq ₃:PVK and Alq ₃:PSS excited states. Finally, the organic light-emitting diode (OLED) structure based on Alq ₃:PVK was fabricated and showed strong electro-luminescence with a green emission at 520 nm. The results of the device show that the ITO/PEDOT:PSS/Alq ₃:PVK/Ca/Al system can be useful for the production of low-cost OLEDs with Alq ₃:PVK as an active layer for future lighting applications.