000			03036nam a22003377a 4500
003			OSt
005			20250716132003.0
008			250716b |||||||| |||| 00| 0 eng d
040			_aTUPM _bEnglish _cTUPM _dTUPM _erda
050			_aBTH TK 870 _bA93 2025
100			_aAvanceña, Angelo SJ. _eauthor
245			_aE-salin: _bdevelopment of a tagalog fsl-to-text and speech-to-text- and-fsl images in real-time web-based filipino sign language and neural machine translation systems for selected 3 major philippine languages: cebuano, ilocano, and waray/ _cAngelo SJ. Avanceña, Rose Belle G. Bolor, Sophia B. Espiritu, Annabela T. Ignacio, Eruel Andrei O. Marasigan, and Aaron B. Mendoza.--
260			_aManila: _bTechnological University of the Philippines, _c2025.
300			_axvii, 272pages: _c29cm.
336			_2rdacontent
337			_2rdamedia
338			_2rdacarrier
500			_aBachelor's thesis
502			_aCollege Of Engineering.-- _bBachelor of science in electronics engineering: _cTechnological University of the Philippines, _d2025.
504			_aIncludes bibliographic references and index.
520			_ae-SALIN is a real-time, web-based translation system designed to bridge the communication gap between the deaf and hearing communities in the Philippines. While many deaf individuals primarily use Filipino Sign Language (FSL), limited understanding among the hearing often leads to miscommunication and exclusion. e-SALIN addresses this by translating FSL into Tagalog text and into three widely spoken Philippine languages: Cebuano, Ilocano, and Waray. It promotes inclusive and respectful communication through gender-fair and culturally sensitive language. Powered by machine learning and computer vision; including CNN and LSTM models. The system accurately recognizes and translates signs in real time, achieving over 94% accuracy within three seconds. It also uses neural machine translation (NMT) models like marianMT and mBART, trained on a dataset of 507 Tagalog words translated into regional languages. Despite working with low-resource languages, it maintains high translation quality, with BLEU scores ranging from 86.9% to 98.8%, highlighting e-SALIN’s strong potential to support inclusive and multilingual communication. A dataset of 507 Tagalog words translated into Cebuano, Ilocano, and Waray using Google Colab Pro + was used to train the marianMT and mBART models. BLEU ratings showed that even with low-resource languages, translation quality varied from 86.9% to 98.8%, indicating good accuracy. These results highlight the system's strong multilingual translation and sign recognition capabilities.
650			_aReal-time system
650			_aWeb-based system
650			_aTranslation system
700			_aBolor, Rose Belle G. _eauthor
700			_aEspiritu, Sophia B. _eauthor
700			_aIgnacio, Annabela T. _eauthor
700			_aMarasigan, Eruel Andrei O. _eauthor
700			_aMendoza, Aaron B. _eauthor
942			_2lcc _cBTH COE _n0
999			_c30416 _d30416