The present study examined the developmental issue of cognitive factors that explain Chinese literacy. Phonological awareness, rapid automatized naming, short-term memory, orthographic awareness and morphological awareness and two literacy tasks (character naming and reading fluency) were administered to 408 second-graders, 428 fourth-graders and 496 six-graders. Results from linear regression analysis and path analysis model showed that the five reading-related cognitive constructs explained unique variances in character naming. Second, character naming is primary for reading fluency after controlling other cognitive constructs; third, the relation between the cognitive factors and literacy changes significantly as a function of reading skills. Results give a clear direction to understanding Chinese reading development.
Excess manganese (Mn) in brain can be neurotoxic, implicated in several neurodegenerative disorders such as sporadic Alzheimer's disease (AD). However, little is known about the altered metal environment including elevated Mn in the progressive cognitive impairment of AD. Indeed, whether high Mn is associated with AD risk remains elusive. In the study, we recruited 40 Chinese elders with different cognitive statuses and investigated concentrations of Mn in whole blood and plasma amyloid-β (Aβ) peptides. Surprisingly, there were significant correlations of Mn with Mini-Mental State Examination score and Clinical Dementia Rating Scale score. In addition, plasma Aβ peptides increased with elevated Mn. Further studies both in vitro and in vivo demonstrated dose-related neurotoxicity and increase of Aβ by Mn treatment, which was probably caused by disrupted Aβ degradation. These data suggested that high Mn may be involved in the progress of AD as an essential pathogenic factor.
Targeting at both high efficiency and performance, we propose AlignTTS to predict the mel-spectrum in parallel. AlignTTS is based on a Feed-Forward Transformer which generates mel-spectrum from a sequence of characters, and the duration of each character is determined by a duration predictor. Instead of adopting the attention mechanism in Transformer TTS to align text to mel-spectrum, the alignment loss is presented to consider all possible alignments in training by use of dynamic programming. Experiments on the LJSpeech dataset show that our model achieves not only state-of-the-art performance which outperforms Transformer TTS by 0.03 in mean option score (MOS), but also a high efficiency which is more than 50 times faster than real-time.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.