Transformer Based Sign-to-Text Translation For Bangladeshi Sign Language

preprint OA: closed CC-BY-4.0
📄 Open PDF Full text JSON View at publisher

Abstract

Abstract Effective communication is a fundamental human need, yet millions of deaf individuals worldwide face significant barriers when interacting with the non-deaf population. Our research addresses this challenge by developing an innovative system that utilizes Natural Language Processing (NLP) techniques to interpret and translate sign language into a format easily understood by individuals who are deaf or hard of hearing. Building on the legacy of assistive technologies, such as the pioneering work in 1977 that enabled the translation of sign language into English via a mechanical hand, our approach aims to modernize and expand these efforts. We have developed a Custom Transformer Encoder Stack Model, trained on a dataset comprising 102 Bangladeshi Sign Language (BdSL) signs. The model achieved exceptional results, with 99.52% accuracy on the training dataset and 98.12% on the test dataset, demonstrating its precision and robustness. Furthermore, the system's adaptability allows for integrating other sign languages through updates to the dataset, making it a flexible and scalable solution for real-time communication. Our research aims to foster greater inclusivity by providing a tool that empowers deaf individuals to communicate seamlessly with the hearing community, thereby enhancing accessibility, social integration, and overall communication equity.
Full text 11,978 characters · extracted from preprint-html · click to expand
Transformer Based Sign-to-Text Translation For Bangladeshi Sign Language | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Transformer Based Sign-to-Text Translation For Bangladeshi Sign Language Tahsinul Haque Dhrubo, Md. Tanzim Reza, Sihab Sahariar, Wasif Afnan Mukto This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7529356/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 06 Dec, 2025 Read the published version in Scientific Reports → Version 1 posted 3 You are reading this latest preprint version Abstract Effective communication is a fundamental human need, yet millions of deaf individuals worldwide face significant barriers when interacting with the non-deaf population. Our research addresses this challenge by developing an innovative system that utilizes Natural Language Processing (NLP) techniques to interpret and translate sign language into a format easily understood by individuals who are deaf or hard of hearing. Building on the legacy of assistive technologies, such as the pioneering work in 1977 that enabled the translation of sign language into English via a mechanical hand, our approach aims to modernize and expand these efforts. We have developed a Custom Transformer Encoder Stack Model, trained on a dataset comprising 102 Bangladeshi Sign Language (BdSL) signs. The model achieved exceptional results, with 99.52% accuracy on the training dataset and 98.12% on the test dataset, demonstrating its precision and robustness. Furthermore, the system's adaptability allows for integrating other sign languages through updates to the dataset, making it a flexible and scalable solution for real-time communication. Our research aims to foster greater inclusivity by providing a tool that empowers deaf individuals to communicate seamlessly with the hearing community, thereby enhancing accessibility, social integration, and overall communication equity. Physical sciences/Engineering Physical sciences/Mathematics and computing Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 06 Dec, 2025 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Revision requested 07 Oct, 2025 Submission checks completed at journal 05 Sep, 2025 First submitted to journal 05 Sep, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7529356","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":510805037,"identity":"95e9b9d9-ab6d-4679-90ff-2a7b420f6341","order_by":0,"name":"Tahsinul Haque Dhrubo","email":"data:image/png;base64,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","orcid":"","institution":"BRAC University","correspondingAuthor":true,"prefix":"","firstName":"Tahsinul","middleName":"Haque","lastName":"Dhrubo","suffix":""},{"id":510805039,"identity":"f12294a9-c44c-493d-b2cd-f1139dc33dbb","order_by":1,"name":"Md. Tanzim Reza","email":"","orcid":"","institution":"BRAC University","correspondingAuthor":false,"prefix":"","firstName":"Md.","middleName":"Tanzim","lastName":"Reza","suffix":""},{"id":510805041,"identity":"e77c52ba-498c-4439-bdf5-41da7e2532e9","order_by":2,"name":"Sihab Sahariar","email":"","orcid":"","institution":"BRAC University","correspondingAuthor":false,"prefix":"","firstName":"Sihab","middleName":"","lastName":"Sahariar","suffix":""},{"id":510805042,"identity":"0d57bbb9-76e3-4a8e-bfea-2d902387ec7b","order_by":3,"name":"Wasif Afnan Mukto","email":"","orcid":"","institution":"BRAC University","correspondingAuthor":false,"prefix":"","firstName":"Wasif","middleName":"Afnan","lastName":"Mukto","suffix":""}],"badges":[],"createdAt":"2025-09-03 17:23:20","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7529356/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7529356/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-025-30856-y","type":"published","date":"2025-12-06T15:57:35+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":97723865,"identity":"c5a8b17a-62e8-438f-98fe-7aa8cc2225df","added_by":"auto","created_at":"2025-12-08 16:08:50","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1205181,"visible":true,"origin":"","legend":"","description":"","filename":"TransformerbasedsigntotexttranslationforBangladeshiSignLanguage.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7529356/v1_covered_1b7a7f64-df72-4db2-a777-bc723d07e7e0.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Transformer Based Sign-to-Text Translation For Bangladeshi Sign Language","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-7529356/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7529356/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Effective communication is a fundamental human need, yet millions of deaf individuals worldwide face significant barriers when interacting with the non-deaf population. Our research addresses this challenge by developing an innovative system that utilizes Natural Language Processing (NLP) techniques to interpret and translate sign language into a format easily understood by individuals who are deaf or hard of hearing. Building on the legacy of assistive technologies, such as the pioneering work in 1977 that enabled the translation of sign language into English via a mechanical hand, our approach aims to modernize and expand these efforts. We have developed a Custom Transformer Encoder Stack Model, trained on a dataset comprising 102 Bangladeshi Sign Language (BdSL) signs. The model achieved exceptional results, with 99.52% accuracy on the training dataset and 98.12% on the test dataset, demonstrating its precision and robustness. Furthermore, the system's adaptability allows for integrating other sign languages through updates to the dataset, making it a flexible and scalable solution for real-time communication. Our research aims to foster greater inclusivity by providing a tool that empowers deaf individuals to communicate seamlessly with the hearing community, thereby enhancing accessibility, social integration, and overall communication equity.","manuscriptTitle":"Transformer Based Sign-to-Text Translation For Bangladeshi Sign Language","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-08 21:41:15","doi":"10.21203/rs.3.rs-7529356/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-10-07T06:01:36+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-09-05T13:00:44+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-09-05T12:57:45+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"f0d368f3-8e6c-44dd-b742-a0ed271852eb","owner":[],"postedDate":"September 8th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":54262069,"name":"Physical sciences/Engineering"},{"id":54262070,"name":"Physical sciences/Mathematics and computing"}],"tags":[],"updatedAt":"2025-12-08T16:01:23+00:00","versionOfRecord":{"articleIdentity":"rs-7529356","link":"https://doi.org/10.1038/s41598-025-30856-y","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2025-12-06 15:57:35","publishedOnDateReadable":"December 6th, 2025"},"versionCreatedAt":"2025-09-08 21:41:15","video":"","vorDoi":"10.1038/s41598-025-30856-y","vorDoiUrl":"https://doi.org/10.1038/s41598-025-30856-y","workflowStages":[]},"version":"v1","identity":"rs-7529356","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7529356","identity":"rs-7529356","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}

Text is read by the "Ask this paper" AI Q&A widget below. Extraction quality varies by source — PMC NXML preserves structure cleanly, OA-HTML may include some navigation residue, and OA-PDF can have broken hyphenation. The publisher copy (via DOI) is the canonical version.

My notes (saved in your browser only)

Ask this paper AI returns verbatim quotes from the full text · source: preprint-html

Answers must be backed by verbatim quotes from this paper's full text. Hallucinated quotes are dropped automatically; if no verbatim passage answers the question, we say so. How this works

Citation neighborhood (no data yet)

We don't have any in-corpus citations linked to this paper yet. This is a recent paper (2025) — citers typically take a year or two to land, and the OpenAlex reference graph may still be filling in.

Source provenance

europepmc
last seen: 2026-05-20T01:45:00.602351+00:00
unpaywall
last seen: 2026-05-29T02:00:03.542394+00:00
License: CC-BY-4.0