Search

The process of obtaining reference patterns for syllablelike units is tedious, error prone, and time consuming. As such, speech recognition systems based on such units are usually tested on only a single talker. In this paper we describe a procedure for using demisyllable reference patterns, excised from spoken utterances for one talker and automatically creating demisyllable reference patterns for a new talker. The procedure is based on dynamic time warping alignment of the spoken utterances (each containing the relevant demisyllable), and the assumption that the optimum warping path identifies the best matching demisyllable within the utterance. The automatic procedure has been used to create demisyllable reference patterns for two new talkers, each talking over a dialed-up telephone line (the original recordings were made over a high-quality microphone). Reference patterns were made for 100 isolated words from a given lexical specification of each word in terms of demisyllables in the inventory. Word recognition accuracies greater than 90% were obtained for both talkers on the 100-word vocabulary. In a second experiment, using a 1109-word vocabulary, recognition accuracies from reference patterns based on automatically extracted demisyllables were from 2%-5% worse than accuracies from reference patterns based on hand corrections applied to the demisyllables. These results show that the automatic demisyllable extraction technique provides a very good first pass set of demisyllables, and that combined with some manual corrections, provides a set of demisyllable patterns suitable for use in a recognition system. [ABSTRACT FROM AUTHOR]

The interpretation of sharp force fatality dynamics may be difficult in some cases, but a contribution to analysis of the phenomenon may be provided by case studies. Therefore, the purpose of our study is focused on identifying, in observed sharp force fatalities, reliable parameters that can differentiate a homicidal and suicidal manner of death, with particular reference to criminological parameters. Data derived from sharp force fatality cases in Padua and Venice from 1997 to 2019, anonymized and collected in Excel, included personal, circumstantial, clinical, and psychopathological-criminological data, as well as crime scene investigation, necroscopic, and toxicological data. Statistical analyses were performed using chi-square and Wilcoxon rank-sum tests. Possible predictors of homicide were analyzed by logistic regression. Six parameters (bloodstains distant from the body, clothing lacerations, hesitation/defense wounds, number of injuries, and potential motives) were significantly different in the two groups (p < 0.05). An independent statistical association between potential motives explaining the crime (p < 0.001; OR 27.533) and homicide on multiple logistic regression analysis was highlighted. The absence of clothing lacerations was inversely related to homicide (p = 0.002, OR 0.092). To the best of our knowledge, this is one of very few Italian studies concerning the differential diagnosis between homicidal and suicidal sharp force fatalities. The dynamics of the event is established in most cases by the integrated evaluation of data from crime scene investigation and the autopsy. Nevertheless, in an atypical scenario, a psychopathological-criminological analysis may provide essential elements, and particular attention should be given to the identification of potential explanatory motives., (© 2020 American Academy of Forensic Sciences.)