scholarly journals Las tecnologías de Reconocimiento Automático de Voz y su incorporación a los métodos de transcripción de lenguas indígenas

2021 ◽  
Vol 55 (2) ◽  
pp. 13
Author(s):  
Hilaria Cruz
Keyword(s):  
Dartmouth College

<p>En agosto del 2018, un retiro en Quechee, Vermont, reunió a lingüistas, hablantes de idiomas en alto riesgo de desaparición, científicos de la computación especializados en procesamiento del lenguaje natural y activistas en el área, con el propósito de discutir las posibilidades de conjuntar esfuerzos para integrar las tecnologías de reconocimiento automático de voz (especialmente las redes neuronales artificiales), a los métodos de transcripción de estas lenguas.</p><p>   En un ambiente ameno, donde el trabajo se mezcló con la diversión, los participantes tuvieron la oportunidad de conocerse, intercambiar ideas, conocimientos y experiencias, dialogando sobre los recursos que el trabajo de documentación lingüística y el reconocimiento de voz, podrían aportar para llevar a cabo dicha meta. Ambos campos compartieron sus últimos avances y las condiciones de sus respectivas áreas de investigación, incluyendo las condiciones de campo, la vitalidad de la lingüística de sus respectivas lenguas, al igual que temas relacionados con el “embotellamiento” de la transcripción del lenguaje.</p><p>   Los lingüistas reportaron sobre su corpus y las horas que habían co-<br /> lectado. De igual manera, los participantes expresaron sus necesidades tecnológicas y sobre cómo debería ser un sistema de Reconocimiento Automático de Voz que pudiera ser usado por personas que no saben mucho sobre tecnología.</p><div><hr align="left" size="1" width="33%" /><div><p><a title="" href="#_ftnref1">[1]</a> Agradezco a la Fundación Neukom, el Programa de Lingüística y el Programa de nativos americanos en el <em>Dartmouth College</em> por apoyar el retiro. Del mismo modo, muchas gracias a Michael Abramov, Oliver Adams y Andrés Perez Perez por sus comentarios en este artículo.</p></div></div>

Interactive Data Analysis of the 1980 Election

1983 ◽  
Vol 38 ◽  
pp. 2-3
Author(s):  
Jon A. Christopherson ◽  
Phillip J. Heyl
Keyword(s):  
Data Analysis ◽  
Voting Behavior ◽  
Social Science Research ◽  
Science Research ◽  
Coast Guard ◽  
Dartmouth College ◽  
Time Sharing ◽  
Batch Mode ◽  
Interactive Data Analysis ◽  
Interactive Data

By introducing data analysis techniques, analyzing current research in voting behavior, and using the computer for social science research, introductory American Government courses on the undergraduate level are undertaking ambitious goals. Using the SETUPS units for each presidential election since 1972, cadets at the Coast Guard Academy have engaged in a comprehensive approach to the study of American electoral behavior in an interactive data analysis mode. Although SETUPS modules are run at most colleges and universities in batch mode, there are imitations and drawbacks to batch mode. The Coast Guard Academy is able to utilize interactive data analysis as a result of its time-sharing arrangement with the Dartmouth College Time Sharing System (DCTS). In this article we will discuss the use of SETUPS in an interactive mode using the CIDAR interactive data analysis system.

Dartmouth College: The BRS Link

1983 ◽  
Vol 1 (1) ◽  
pp. 55-58
Author(s):  
Emily Fayen
Keyword(s):  
Dartmouth College

scholarly journals Can helmet design reduce the risk of concussion in football?

2014 ◽  
Vol 120 (4) ◽  
pp. 919-922 ◽  
Author(s):  
Steven Rowson ◽  
Stefan M. Duma ◽  
Richard M. Greenwald ◽  
Jonathan G. Beckwith ◽  
Jeffrey J. Chu ◽  
...  
Keyword(s):  
Head Impact ◽  
Dartmouth College ◽  
University Of Minnesota ◽  
Head Impacts ◽  
University Of Oklahoma ◽  
Significant Difference ◽  
Rule Changes ◽  
Impact Data ◽  
Collegiate Football ◽  
Helmet Design

Of all sports, football accounts for the highest incidence of concussion in the US due to the large number of athletes participating and the nature of the sport. While there is general agreement that concussion incidence can be reduced through rule changes and teaching proper tackling technique, there remains debate as to whether helmet design may also reduce the incidence of concussion. A retrospective analysis was performed of head impact data collected from 1833 collegiate football players who were instrumented with helmet-mounted accelerometer arrays for games and practices. Data were collected between 2005 and 2010 from 8 collegiate football teams: Virginia Tech, University of North Carolina, University of Oklahoma, Dartmouth College, Brown University, University of Minnesota, Indiana University, and University of Illinois. Concussion rates were compared between players wearing Riddell VSR4 and Riddell Revolution helmets while controlling for the head impact exposure of each player. A total of 1,281,444 head impacts were recorded, from which 64 concussions were diagnosed. The relative risk of sustaining a concussion in a Revolution helmet compared with a VSR4 helmet was 46.1% (95% CI 28.1%–75.8%). When controlling for each player's exposure to head impact, a significant difference was found between concussion rates for players in VSR4 and Revolution helmets (χ2 = 4.68, p = 0.0305). This study illustrates that differences in the ability to reduce concussion risk exist between helmet models in football. Although helmet design may never prevent all concussions from occurring in football, evidence illustrates that it can reduce the incidence of this injury.

Sign in / Sign up

Export Citation Format

Share Document