Search

With the aim of adapting a source Text to Speech (TTS) model to synthesize a personal voice by using a few speech samples from the target speaker, voice cloning provides a specific TTS service. Although the Tacotron 2-based multi-speaker TTS system can implement voice cloning by introducing a d-vector into the speaker encoder, the speaker characteristics described by the d-vector cannot allow for the voice information of the entire utterance. This affects the similarity of voice cloning. As a vocoder, WaveNet sacrifices speech generation speed. To balance the relationship between model parameters, inference speed, and voice quality, a voice cloning method based on improved HiFi-GAN has been proposed in this paper. (1) To improve the feature representation ability of the speaker encoder, the x-vector is used as the embedding vector that can characterize the target speaker. (2) To improve the performance of the HiFi-GAN vocoder, the input Mel spectrum is processed by a competitive multiscale convolution strategy. (3) The one-dimensional depth-wise separable convolution is used to replace all standard one-dimensional convolutions, significantly reducing the model parameters and increasing the inference speed. The improved HiFi-GAN model remarkably reduces the number of vocoder model parameters by about 68.58% and boosts the model's inference speed. The inference speed on the GPU and CPU has increased by 11.84% and 30.99%, respectively. Voice quality has also been marginally improved as MOS increased by 0.13 and PESQ increased by 0.11. The improved HiFi-GAN model exhibits outstanding performance and remarkable compatibility in the voice cloning task. Combined with the x-vector embedding, the proposed model achieves the highest score of all the models and test sets. [ABSTRACT FROM AUTHOR]

Aiming at the operation and maintenance requirements of the fault location of high-temperature superconducting cables, a fault location method of high-temperature superconducting cables based on the improved time-frequency domain reflection method and EEMD noise reduction is proposed. Considering the cross-term interference problem in the traditional time-frequency domain reflection method, this paper introduces the affine transformation to project the time-frequency distribution of the self-term and the cross term and further highlights the characteristic differences between the two through coordinate transformation, and the particle swarm algorithm is employed to solve the optimal stagger angle of the affine transformation. The unscented particle filter is adopted to separate the cross term, and EEMD noise reduction is introduced to solve the signal noise problem. Finally, two software programs, PSCAD and MATLAB, are employed for joint simulation to build a model of high-temperature superconducting cable. The simulation example shows that the proposed method in this paper can eliminate the cross-term interference of the traditional time-frequency domain reflection method, effectively locate the fault of the high-temperature superconducting cable, and improve the positioning accuracy. [ABSTRACT FROM AUTHOR]

The stability of the reservoir bank landslide is affected by a variety of external factors, and the fluctuation of reservoir water level is one of the important influencing factors. The Erdaohe landslide is a typically colluvial landslide in the Three Gorges Reservoir area with periodic reservoir water level fluctuations. According to landslide displacement data, the displacement of the Erdaohe landslide exhibits the significantly stepwise feature. Its failure mechanism was analyzed using strength reduction method by the FLAC3D package in the case of reservoir water level changes. The results indicate that the hydrodynamic pressure has an important impact on the initialization of the landslide failure. When reservoir water level rises rapidly or maintains constant at the lower level, the landslide stability would be higher. When the reservoir water level decreases rapidly or maintains constant at the higher level, the landslide stability will be smaller. When the reservoir water level was in the lowest elevation, the factor of safety (FS) reached the minimum value of 1.11. Findings in this paper can provide guidelines for the risk assessment of colluvial landslides. [ABSTRACT FROM AUTHOR]

The large-scale rectangular pipe jacking technology has attracted more and more attention in the construction of urban underground pipe galleries, especially the problem of prediction and control of the surface settlement. Taking the pipe gallery project of Chengbei Road in Suzhou as an example, the settlement values of three groups of monitoring sections under the conditions of natural settlement and synchronous grouting are monitored, and the measured settlement values are compared with Peck formula and random medium theory. The results show that the Peck formula is more accurate in predicting the natural settlement within 1.5 times the pipe jacking width, while the stochastic medium theory is more accurate in predicting the settlement beyond the width. By comparing the field test and theory, the suggestions on the prediction of ground settlement of rectangular pipe jacking of large scale are put forward. The results of this study provide a basis for subsequent related research work and engineering applications. [ABSTRACT FROM AUTHOR]

With the deepening of tourists' demand for tourism services, the personalization of online tourists has gradually become an application of personalized recommendation technology. According to the application requirements of personalized scenic spot recommendation, this paper uses social networks and Bayesian networks to fully mine the matching degree between users and scenic spots for personalized recommendation. Add social network factors to the recommendation of tourist attractions, and fully tap the social network relationship between users. First, the users are clustered by the coupling bidirectional clustering algorithm. Then, DBSCAN (density-based noise application spatial clustering) algorithm is used to cluster scenic spots. Finally, two stable user sets and scenic spot sets are applied to the personalized recommendation algorithm to predict the user's next upcoming scenic spot. The algorithm is compared with some traditional algorithms in the dataset. The algorithm deals with the similarity of user attributes and user behavior and uses content-based algorithm to deal with the relationship between scenic spots; com datasets have better performance. [ABSTRACT FROM AUTHOR]

Horizontal wells are extensively utilized in the development of unconventional reservoirs. However, the logging responses and formation evaluation in horizontal wells can be impacted by factors like anisotropy and tool eccentricity. To investigate the influence of tool eccentricity on acoustic logging response, physical simulation experiments of array acoustic logging were conducted in a scaled borehole formation model under different tool eccentricity conditions. The experimental data were analyzed, and the findings revealed that when the receiver array is parallel to the borehole axis, the P-wave slowness and S-wave slowness remain unaffected by tool eccentricity. However, the amplitudes of the P-wave and S-wave decrease significantly with increasing tool eccentricity, following an approximate negative exponential pattern. Additionally, when the transmitter is centered and the receiver array intersects the borehole axis at an angle, the wave velocities increase significantly with tool eccentricity, with the P-wave velocity showing a faster increase. Conversely, when the transmitter is eccentric and the receiver array intersects the borehole axis at an angle, the wave velocity decreases notably with tool eccentricity, and the P-wave velocity decreases even faster. These findings contribute to a better understanding of the impact of tool eccentricity on array acoustic logging response in horizontal wells and offer guidance for developing correction schemes to address this effect. [ABSTRACT FROM AUTHOR]

In order to study the volatile and semivolatile components changes of cut tobacco in the drying process, the cut tobacco were determined by headspace-gas chromatography/mass spectrometry (HS-GC/MS). Principal component analysis (PCA) and similarity analysis were used to analyze the HS-GC/MS results. Based on the similarity, the drying conditions and drying mechanism have been studied. The results showed that the increase in water content would reduce the dying rate, and if the temperature increased, the dying rate increased. The preliminary study on the drying kinetics of the drying process shows that Henderson and Pabis model appeared to be the most suitable for describing the drying process of cut tobacco, and the drying activation energy is 111.9 kJ. [ABSTRACT FROM AUTHOR]

This study was designed to investigate the effects of nonpharmacological integrated care protocols on fatigue in patients with hemodialysis. This parallel randomized controlled trial was conducted on patients undergoing hemodialysis from May to October 2020 at the Dialysis Center of the Fifth Affiliated Hospital of Zunyi Medical University. The patients were randomized into an intervention group (accepting nonpharmacological integrated care protocols and standard care) or a control group (accepting standard care only) using a computer-generated random number. The nonpharmacological holistic care intervention used in this study involved a well-rounded multidisciplinary team that worked together to improve dietary compliance, medication adherence, and self-management to improve patients' care and promote self-management. From the 120 evaluated patients, 116 cases were eligible and analyzed. The results showed that patients from the intervention group had obviously reduced overall fatigue, mental fatigue, and muscular fatigue relative to the control group. The nonpharmacological integrated care protocols were interactive and promotive to each other. Meanwhile, the role and function of nurses in the management of chronic disease were demonstrated to be crucial. [ABSTRACT FROM AUTHOR]

In the field of materials research, scanning microwave microscopy imaging has already become a vital research tool due to its high sensitivity and nondestructive testing of samples. In this article, we review the main theoretical and fundamental components of microwave imaging, in addition to the wide range of applications of microwave imaging. Rather than the indirect determination of material properties by measuring dielectric constants and conductivity, microwave microscopy now permits the direct investigation of semiconductor devices, electromagnetic fields, and ferroelectric domains. This paper reviews recent advances in scanning microwave microscopy in the areas of resolution and operating frequency and presents a discussion of possible future industrial and academic applications. [ABSTRACT FROM AUTHOR]

Hepatocellular carcinoma (HCC) is a notoriously difficult cancer to treat. The recent development of immune checkpoint inhibitors has revolutionised HCC therapy; however, successful response is only observed in a small percentage of patients. Biomarkers typically used to predict treatment response in other tumour types are ineffective in HCC, which arises in an immune-suppressive environment. However, imaging markers that measure changes in tumour infiltrating immune cells may supply information that can be used to determine which patients are responding to therapy posttreatment. We have evaluated [18F]AlF-mNOTA-GZP, a radiolabeled peptide targeting granzyme B, to stratify response to ICIs in a HEPA 1-tumours, a syngeneic model of HCC. Posttherapy, in vivo tumour retention of [18F]AlF-mNOTA-GZP was correlated to changes in tumour volume and tumour-infiltrating immune cells. [18F]AlF-mNOTA-GZP successfully stratified response to immune checkpoint inhibition in the syngeneic HEPA 1-6 model. FACS indicated significant changes in the immune environment including a decrease in immune suppressive CD4+ T regulatory cells and increases in tumour-associated GZB+ NK+ cells, which correlated well with tumour radiopharmaceutical uptake. While the immune response to ICI therapies differs in HCC compared to many other cancers, [18F]AlF-mNOTA-GZP retention is able to stratify response to ICI therapy associated with tumour infiltrating GZB+ NK+ cells in this complex tumour microenvironment. [ABSTRACT FROM AUTHOR]

Hepatocellular carcinoma (HCC) is a notoriously difficult cancer to treat. The recent development of immune checkpoint inhibitors has revolutionised HCC therapy; however, successful response is only observed in a small percentage of patients. Biomarkers typically used to predict treatment response in other tumour types are ineffective in HCC, which arises in an immune-suppressive environment. However, imaging markers that measure changes in tumour infiltrating immune cells may supply information that can be used to determine which patients are responding to therapy posttreatment. We have evaluated [18F]AlF-mNOTA-GZP, a radiolabeled peptide targeting granzyme B, to stratify response to ICIs in a HEPA 1-tumours, a syngeneic model of HCC. Posttherapy, in vivo tumour retention of [18F]AlF-mNOTA-GZP was correlated to changes in tumour volume and tumour-infiltrating immune cells. [18F]AlF-mNOTA-GZP successfully stratified response to immune checkpoint inhibition in the syngeneic HEPA 1-6 model. FACS indicated significant changes in the immune environment including a decrease in immune suppressive CD4+ T regulatory cells and increases in tumour-associated GZB+ NK+ cells, which correlated well with tumour radiopharmaceutical uptake. While the immune response to ICI therapies differs in HCC compared to many other cancers, [18F]AlF-mNOTA-GZP retention is able to stratify response to ICI therapy associated with tumour infiltrating GZB+ NK+ cells in this complex tumour microenvironment. [ABSTRACT FROM AUTHOR]