| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 4.61 MB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
O setor da saúde e da tecnologia têm vindo a ter um grande avanço nas últimas décadas, contudo, ainda existe muitos obstáculos a ultrapassar no que diz respeito à adaptação da sociedade às pessoas com deficiência, nomeadamente a deficiência auditiva. Em Portugal, existem entre 100 a 150 mil pessoas com algum nível de perda auditiva e dessas, cerca de 30 mil utilizam Língua Gestual Portuguesa como língua materna. As maiores dificuldades encontradas pelas pessoas com deficiência auditiva são a pobre comunicação com as pessoas ouvintes, comunicação esta, que se tornou ainda mais complicada com a obrigatoriedade do uso de máscara facial devido à Pandemia Covid-19. Esta dissertação teve como foco o desenvolvimento de um sistema de tradução automático de voz e texto para Língua Gestual Portuguesa. Para tal, foi necessário recorrer a diferentes ferramentas a nível de linguagens de programação de modo a criar um tradutor que contém 23 imagens na base de dados e um grau de confiança entre os 70 e os 90%. Este tradutor difere dos restantes tradutores automáticos disponíveis, uma vez que não utiliza avatares para gestuar, tendo sido criado um tradutor exclusivamente por imagens. O sistema de tradução foi avaliado através de um questionário feito por 3 pessoas surdas da Associação de Surdos do Concelho de Sintra, que utilizam a LGP como meio de comunicação no seu quotidiano, onde avaliaram que o tradutor tem um bom desempenho e será útil para comunidade surda.
ABSTRACT - The health and technology sectors have made great advances in the last decades, however, there are still many obstacles to overcome regarding the adaptation of society to people with disabilities, namely hearing impairment. In Portugal, there are between 100 to 150 thousand people with some level of hearing loss, and of these, around 30 thousand use Portuguese Sign Language as their mother tongue. The greatest difficulty encountered by the deaf and hearing impaired is poor communication with hearing people, communication which became even more complicated with the mandatory use of face masks due to the COVID-19 pandemic. This dissertation focused on the development of an automatic voice and text translation system for Portuguese Sign Language. For that, it was necessary to use different programming language tools to create a translator with 23 images in the database and a confidence level between 70 and 90%. This translator differs from the other automatic translators available since it does not use avatars to gesture, and a translator exclusively by images has been created. The translation system was evaluated through a questionnaire made by three deaf people from the Associação de Surdos do Concelho de Sintra, who use LGP as a means of communication in their daily lives. They evaluated that the translator has a good performance and will be useful for the deaf community.
ABSTRACT - The health and technology sectors have made great advances in the last decades, however, there are still many obstacles to overcome regarding the adaptation of society to people with disabilities, namely hearing impairment. In Portugal, there are between 100 to 150 thousand people with some level of hearing loss, and of these, around 30 thousand use Portuguese Sign Language as their mother tongue. The greatest difficulty encountered by the deaf and hearing impaired is poor communication with hearing people, communication which became even more complicated with the mandatory use of face masks due to the COVID-19 pandemic. This dissertation focused on the development of an automatic voice and text translation system for Portuguese Sign Language. For that, it was necessary to use different programming language tools to create a translator with 23 images in the database and a confidence level between 70 and 90%. This translator differs from the other automatic translators available since it does not use avatars to gesture, and a translator exclusively by images has been created. The translation system was evaluated through a questionnaire made by three deaf people from the Associação de Surdos do Concelho de Sintra, who use LGP as a means of communication in their daily lives. They evaluated that the translator has a good performance and will be useful for the deaf community.
Description
Mestrado em Engenharia Biomédica
Keywords
Deficiência auditiva Comunicação COVID-19 Língua gestual portuguesa Tradutor automático Hearing impaired Communication Portuguese sign language Automatic translator
Citation
Brito M. Tradutor automático de língua portuguesa (voz e texto) para língua gestual portuguesa [dissertation]. Lisboa: Escola Superior de Tecnologia da Saúde de Lisboa- Instituto Superior de Engenharia de Lisboa/Instituto Politécnico de Lisboa; 2022.
Publisher
Instituto Politécnico de Lisboa, Escola Superior de Tecnologia da Saúde de Lisboa