We describe an online handwriting system that is able to support 102 languages using a deep neural network architecture. This new system has completely replaced our previous segment-and-decode-based system and reduced the error rate by 20-40% relative for most languages. Further, we report new state-of-the-art results on IAM-OnDB for both the open and closed dataset setting. The system combines methods from sequence recognition with a new input encoding using Bézier curves. This leads to up to 10× faster recognition times compared to our previous system. Through a series of experiments, we determine the optimal configuration of our models and report the results of our setup on a number of additional public datasets.

A COVID-19 chegou rapidamente na periferia das grandes cidades brasileiras, socialmente mais vulnerável. A baixa capacidade de testagem resulta na adoção de medidas sem informações consistentes sobre o comportamento da doença e interfere na adoção de ações de controle. Objetivos: estimar a prevalência de infecção por SARS-CoV-2 na população da Região Metropolitana da Baixada Santista (RMBS) e analisar impactos da vulnerabilidade social e de políticas públicas implementadas em contextos de desigualdades. Métodos: estudo de caráter quantitativo, transversal, através de inquérito sorológico seriado e aplicação de questionário em amostragem populacional estratificada e coleta domiciliar, nos nove municípios da RMBS. Conclusões: A soroprevalência medida foi de 1,4% na primeira fase e de 2,2 % na segunda permitindo estimar 15 pessoas infectadas para cada caso notificado na primeira fase, e 10 na seguinte. A letalidade foi recalculada para 0,40% e 0,48 % em cada fase, aproximando-se da casuística internacional. Pessoas mais vulneráveis são as mais atingidas pela pandemia, devendo ser consideradas: a informalidade no trabalho, baixa renda, cor da pele auto referida como preta ou parda e informações ambivalentes quanto à prevenção. Os resultados reforçam a importância do isolamento social e de adoção de medidas econômicas e sociais protetivas destinadas às populações vulneráveis.

Calvo Lance, M.; Hurtado Oliver, LF.; García Granada, F.; Sanchís Arnal, E. (2012) Abstract. In this paper, we present a statistical approach to Language Understanding that allows to avoid the effort of obtaining new semantic models when changing the language. This way, it is not necessary to acquire and label new training corpora in the new language. Our approach consists of learning all the semantic models in a target language and to do the semantic decoding of the sentences pronounced in the source language after a translation process. In order to deal with the errors and the lack of coverage of the translations, a mechanism to generalize the result of several translators is proposed. The graph of words generated in this phase is the input to the semantic decoding algorithm specifically designed to combine statistical models and graphs of words. Some experiments that show the good behavior of the proposed approach are also presented.

Abstract. In this paper, we present an approach to Spoken Language Understanding, where the input to the semantic decoding process is a composition of multiple hypotheses provided by the Automatic Speech Recognition module. This way, the semantic constraints can be applied not only to a unique hypothesis, but also to other hypotheses that could represent a better recognition of the utterance. To do this, we have developed an algorithm to combine multiple sentences into a weighted graph of words, which is the input to the semantic decoding process. It has also been necessary to develop a specific algorithm to process these graphs of words according to the statistical models that represent the semantics of the task. This approach has been evaluated in a SLU task in Spanish. Results, considering different configurations of ASR outputs, show the better behavior of the system when a combination of hypotheses is considered.

Background: COVID-19 reached Brazil in February 2020 after spreading throughout Asia and Europe. The disease first struck in Brazil major cities, affecting high-income groups, subsequently spreading to the socially-vulnerable outlying areas. The objective of this preliminary study was to the estimate the seroprevalence of COVID-19 in the Baixada Santista Metropolitan Area - RMBS, to inform policies for prevention and preparation of primary care, hospital services, hospital beds, intensive and critical care in a bid to reduce lethality. Methods: This study is the first of a four-phase cross-sectional survey to include all cities within the Baixada Santista metropolitan area. The probabilistic population-based sample was based on a 7% prevalence with a 2% delta, a 5% level of significance and 90% power, and considered the estimated population, stratified by age, gender and city. The sample was randomized in each city and street selection was based on data drawn from the 2010 Brazilian census. The serology test was chosen from the available test kits approved by the National Health Agency (ANS) of the National Health Surveillance Agency (ANVISA), for the detection of SARS-CoV-2 antibodies. Information on socioeconomic characteristics, health service use, symptoms and adherence to social distancing measures was collected. All variables were expressed as absolute and relative frequencies and confidence intervals. Results: The final sample comprised 2,342 respondents with a mean age of 37.78 (19.98) years and equal gender distribution. Respondents predominantly lived in an abode with 4 or more rooms, 71.9% had an income of ≤3 minimum wages and no formal employment. Over half of the sample reported being reliant on the National Health Service (SUS) for health care, while 26.5% held private health insurance. Social distancing was practiced by 61.4% of respondents who answered this question. Seroprevalence measured was 1.4% for a 95% CI (0.93-1.93) based on mathematical estimates, with a ratio of 15 actual infections for every case notified, and lethality of 0.48%. Conclusion: Results revealed a flattening of the epidemiological curve in the region and high underreporting of cases.