Nuclear Magnetic Resonance (NMR) spectroscopy is a cornerstone technique for determining the structures of small molecules and is especially critical in the discovery of novel natural products and clinical therapeutics. Yet, interpreting NMR spectra remains a time-consuming, manual process requiring extensive domain expertise. We introduce ChefNMR (CHemical Elucidation From NMR), an end-to-end framework that directly predicts an unknown molecule's structure solely from its 1D NMR spectra and chemical formula. We frame structure elucidation as conditional generation from an atomic diffusion model built on a non-equivariant transformer architecture. To model the complex chemical groups found in natural products, we generated a dataset of simulated 1D NMR spectra for over 111,000 natural products. ChefNMR predicts the structures of challenging natural product compounds with an unsurpassed accuracy of over 65%. This work takes a significant step toward solving the grand challenge of automating small-molecule structure elucidation and highlights the potential of deep learning in accelerating molecular discovery.

#StopAsianHate and #StopAAPIHate are two of the most commonly used hashtags that represent the current movement to end hate crimes against the Asian American and Pacific Islander communities. We conduct a social media study of public opinion toward the #StopAsianHate and #StopAAPIHate movement based on 46 058 Twitter users across 30 states in USA ranging from March 18, 2021 to April 11, 2021. To facilitate fine-grained analyses, the demographic information of the Twitter users, including age, gender, race/ethnicity, social capital, political affiliation, geolocation, family, income, and religious status, is either retrieved from the user profiles or inferred using classifiers. We find that the movement attracts more participation from women, younger adults, Asian, and Black communities. It is noteworthy that most of them are also active in other online movements related to racial or social issues, such as #BlackLivesMatter or #SayHerName; 51.56% of the Twitter users show direct support, 18.38% are news about anti-Asian hate crimes, and 5.43% show a negative attitude toward the movement. By conducting logistic regression, we find that the public opinion varies across user characteristics. Furthermore, among the states with most racial bias-motivated hate crimes, the negative attitude toward the #StopAsianHate and #StopAAPIHate movement is the weakest. The majority of the top influencers are Asian–American reporters, journalists, or politicians. To the best of our knowledge, this is the first large-scale social media-based study to understand public opinion toward the #StopAsianHate and #StopAAPIHate movement. We hope that our study can provide insights and promote research on anti-Asian hate crimes and ultimately help address such a serious societal issue for the common benefits of all communities.

Vocal aging, a universal process of human aging, can largely affect one's language use, possibly including some subtle acoustic features of one's utterances like Voice Onset Time. To figure out the time effects, Queen Elizabeth's Christmas speeches are documented and analyzed in the long-term trend. We build statistical models of time dependence in Voice Onset Time, controlling a wide range of other fixed factors, to present annual variations and the simulated trajectory. It is revealed that the variation range of Voice Onset Time has been narrowing over fifty years with a slight reduction in the mean value, which, possibly, is an effect of diminishing exertion, resulting from subdued muscle contraction, transcending other non-linguistic factors in forming Voice Onset Time patterns over a long time.

In this study, we categorize the regional accented Chinese into 3 levels of accents according to phonetic annotation and subjective evaluation on a regional accented speech corpus of three regions: Shanghai, Wuhan and Xiamen. Three accent evaluation methods, namely segmental annotation, clustering on phonetic annotation and subjective evaluation, are compared based on phonetic error rates. The results show that objective evaluation score based on segmental pronunciation is higher than subjective evaluation for the same speaker. This implies that supra-segmental features play an important role in rating accent degree and segmental features alone are not enough for objective evaluation. In accent level criterion, the errors from prosodic and segmental aspects are not equal and the percentage of these two parts are various for different regional speakers. The result is helpful for machine evaluation, L2 teaching and acquisition.