On phase recovery and preserving early reflections for deep-learning speech dereverberation.

In indoor environments, reverberation often distorts clean speech. Although deep learning-based speech dereverberation approaches have shown much better performance than traditional ones, the inferior speech quality of the dereverberated speech caused by magnitude distortion and limited phase recovery is still a serious problem for practical applications. This paper improves the performance of deep learning-based speech dereverberation from the perspectives of both network design and mapping target optimization. Specifically, on the one hand, a bifurcated-and-fusion network and its guidance loss functions were designed to help reduce the magnitude distortion while enhancing the phase recovery. On the other hand, the time boundary between the early and late reflections in the mapped speech was investigated, so as to make a balance between the reverberation tailing effect and the difficulty of magnitude/phase recovery. Mathematical derivations were provided to show the rationality of the specially designed loss functions. Geometric illustrations were given to explain the importance of preserving early reflections in reducing the difficulty of phase recovery. Ablation study results confirmed the validity of the proposed network topology and the importance of preserving 20 ms early reflections in the mapped speech. Objective and subjective test results showed that the proposed system outperformed other baselines in the speech dereverberation task.

Sixty Years of Frequency-Domain Monaural Speech Enhancement: From Traditional to Deep Learning Methods.

Frequency-domain monaural speech enhancement has been extensively studied for over 60 years, and a great number of methods have been proposed and applied to many devices. In the last decade, monaural speech enhancement has made tremendous progress with the advent and development of deep learning, and performance using such methods has been greatly improved relative to traditional methods. This survey paper first provides a comprehensive overview of traditional and deep-learning methods for monaural speech enhancement in the frequency domain. The fundamental assumptions of each approach are then summarized and analyzed to clarify their limitations and advantages. A comprehensive evaluation of some typical methods was conducted using the WSJ + Deep Noise Suppression (DNS) challenge and Voice Bank + DEMAND datasets to give an intuitive and unified comparison. The benefits of monaural speech enhancement methods using objective metrics relevant for normal-hearing and hearing-impaired listeners were evaluated. The objective test results showed that compression of the input features was important for simulated normal-hearing listeners but not for simulated hearing-impaired listeners. Potential future research and development topics in monaural speech enhancement are suggested.

Benzoic Acid: Electrode-Regenerated Molecular Catalyst to Boost Cycloolefin Epoxidation.

Stoichiometric oxidants are always consumed in organic oxidation reactions. For example, olefins react with peroxy acids to be converted to epoxy, while the oxidant, peroxy acid, is downgraded to carboxylic acid. In this paper, we aim to regenerate carboxylic acid into peroxy acid through electric water splitting at the anode, in order to construct an electrochemical catalytic cycle to accomplish the cycloolefin epoxidation reaction. Benzoic acid, which can be strongly adsorbed onto the anode and rapidly converted to peroxy acid, was selected to catalyze the cycloolefin epoxidation. Furthermore, the peroxybenzoic acid will be further activated on the electrode to fulfill the epoxidation and release the benzoic acid to complete the catalytic cycle. In this designed reaction cycle, benzoic acid acts as a molecular catalyst with the assistance of the electrode-generated reactive oxygen species (ROS). This method can successfully reform the consumable oxidants to molecular catalysts, which can be generalized to other green organic syntheses.

COVID-19 vaccine uptake and hesitation among men and women preparing for pregnancy: a cross-section survey based on the theory of planned behavior.

BACKGROUND: Given the accelerated speed of COVID-19 vaccine research and administration, the main barriers to herd immunity appear to be concerns about safety and efficacy. Men and women preparing for pregnancy may have the same concerns about COVID-19 vaccination, but few studies have focused on COVID-19 vaccine uptake and hesitation among them. METHODS: A cross-sectional study was conducted among men and women who were preparing for pregnancy in Southwest China. The questionnaire was designed based on the theory of planned behavior (TPB). Multiple logistic regression was used to explore the determinants of the behaviors of COVID-19 vaccination. RESULTS: A total of 2878 participants completed the survey. A total of 53.89% of participants received at least one dose of the COVID-19 vaccine. A total of 45.21% of participants would receive the COVID-19 vaccine in the future. A total of 0.90% of participants never thought about receiving the COVID-19 vaccine. Multiple logistic regression model 1 showed that female participants (OR:5.497, 95%CI: 4.292-7.041), participants who never received influenza vaccine (OR:2.664, 95%CI: 1.908-3.718), participants who had never been tested for COVID-19 (OR:2.244, 95%CI:1.504-3.349), participants who had higher score of negative attitude (OR:1.448, 95%CI: 1.219-1.719), participants who had lower scores of injunctive norms (OR:0.440, 95%CI: 0.360-0.537) and descriptive norms (OR:0.105, 95%CI: 0.088-0.126) were more likely to delay COVID-19 vaccination. Model 2 showed that participants who had lower scores for positive attitude (OR: 0.406, 95% CI: 0.230-0.716), injunctive norms (OR: 0.283, 95% CI: 0.130-0.614) and descriptive norms (OR: 0.060, 95% CI: 0.038-0.094) were more likely to refuse COVID-19 vaccination. CONCLUSIONS: The COVID-19 vaccination rate of men and women preparing for pregnancy was significantly lower than the average vaccination rate of China. Gender, protective health behaviors, vaccination attitudes, and subjective norms had effects on the vaccination behaviors of couples preparing for pregnancy.

Electrochemical oxidation of styrene to benzaldehyde by discrimination of spin-paired π electrons.

The oxidation of styrene to benzaldehyde has been a considerable challenge in the electrochemical synthesis of organic compounds because styrene is more easily oxidized to benzoic acid. In this work, MnO2 with an asymmetric electronic configuration is designed to discriminate the spin-paired π electrons of styrene. One of these discriminated π electrons combined with reactive oxygen species (ROS), ˙OH, ˙OOH, etc., produced simultaneously on a MnO2/(Ru0.3Ti0.7)O2/Ti bifunctional anode, to form benzaldehyde via Grob fragmentation, rather than benzoic acid. However, only benzoic acid is obtained from the oxidation of styrene on the anodes MOs/(Ru0.3Ti0.7)O2/Ti, where MOs are other metal oxides with symmetric electronic configurations.

A thio-ß-cyclodextrin functionalized graphene/gold nanoparticle electrochemical sensor: a study of the size effect of the gold nanoparticles and the determination of tetrabromobisphenol A.

In this study, a novel tetrabromobisphenol A (TBBPA) sensor was fabricated based on a CTAB-capped gold nanoparticle (AuNPs)-thio-ß-cyclodextrin (SH-ß-CD)/graphene oxide modified glassy carbon electrode (GCE). The peak current of TBBPA was dramatically enhanced by the AuNPs with a diameter of 6.2 nm on the modified electrodes compared with the other sized particles (10.1 or 16.1 nm). To further improve the electrochemical performance of the modified electrode, the influence of pH of the buffer solution and the accumulation time on the determination were investigated. The optimum pH and accumulation time were 7.0 and 180 s, respectively. The developed sensor exhibited good reproducibility, and excellent sensitivity and selectivity, showing a low detection limit (1.2 × 10-9 mol L-1) and a linear range from 1.5 × 10-8 to 7 × 10-6 mol L-1. In addition, a possible oxidization mechanism of TBBPA was also discussed. Finally, this sensor was successfully applied to detect TBBPA in water samples, and the results were consistent with those acquired by high-performance liquid chromatography.