<link rel="stylesheet" href="styles.f3b1fba60ec7970c.css">

Development of SIBI and BISINDO Fingerspelling Learning Media Based on Cloud Voice Recognition: An Iterative Development Approach

dc.contributor.authorSukma, Muhammad Fazli Ramadhani
dc.date.accessioned2026-11-14T08:11:44Z
dc.date.issued2026
dc.description.abstractCommunication barriers between the hearing public and the Deaf community remain a significant problem, partly because public knowledge of sign language is limited and conventional static media offer little interactivity for learning fingerspelling. This study develops an Android application as an interactive learning medium that translates spoken words into Indonesian sign language fingerspelling, covering both the SIBI (Sistem Isyarat Bahasa Indonesia) and BISINDO (Bahasa Isyarat Indonesia) standards, using cloud-based voice recognition. The application was developed using the Iterative Development Method comprising analysis, design, implementation, testing, and evaluation phases. To respect Deaf culture, the application is intentionally limited to translating fingerspelling for the letters A to Z. Development proceeded through two architectural scenarios: an initial on-device processing approach (Scenario 1) and a cloud-based approach using the AssemblyAI REST API (Scenario 2). The results show that migrating from on-device processing to the cloud-based architecture substantially reduced transcription latency, from approximately 83 seconds per word to about 3.35 seconds per word on the same device. Black Box Testing confirmed that all application components (state management, user interface, translation practice, dictionary, and evaluation quiz) functioned correctly with a 100% pass rate after iterative refinement, and an end-to-end integration test produced correct fingerspelling visualization for all 45 multi-character word inputs, while single-letter inputs (A–E) were not recognized. User Acceptance Testing with 19 respondents, conducted via Google Form using a five-point Likert scale across four aspects, yielded an overall acceptance score of 89.58%, categorized as Very Feasible (Sangat Layak).
dc.identifier.urihttps://repository.uii.ac.id/handle/123456789/67767
dc.publisherUniversitas Islam Indonesia
dc.subjectLearning Media
dc.subjectSign Language
dc.subjectSIBI
dc.subjectBISINDO
dc.subjectVoice Recognition
dc.subjectCloud ASR
dc.subjectAssemblyAI
dc.subjectAndroid
dc.titleDevelopment of SIBI and BISINDO Fingerspelling Learning Media Based on Cloud Voice Recognition: An Iterative Development Approach
dc.typeThesis

Files

Original bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
22523070.pdf
Size:
1.56 MB
Format:
Adobe Portable Document Format