Indonesian Sign Language API (OpenSIBI API) as The Gateway Services for Myo Armband

<p><em>Hand gestures are the communication ways for the deaf people and the other. Each hand gesture has a different meaning. In order to better communicate, we need an automatic translator who can recognize hand movements as a word or sentence in communicating with deaf people. </em><em>This paper proposes a system to recognize hand gestures based on Indonesian Sign Language Standard. This system uses Myo Armband as hand gesture sensors. Myo Armband has 21 sensors to express the hand gesture data. Recognition process uses a Support Vector Machine (SVM) to classify the hand gesture based on the dataset of Indonesian Sign Language Standard. SVM yields the accuracy of 86.59% to recognize hand gestures as sign language.</em></p><p><em><strong>Keywords</strong></em><em>: </em><em>Hand Gesture Recognition, Feature Extraction, Indonesian Sign Language, Myo Armband, Moment Invariant</em></p>

Myoelectric prostheses are a viable solution for people with amputations. The challenge in implementing a usable myoelectric prosthesis lies in accurately recognizing different hand gestures. The current myoelectric devices usually implement very few hand gestures. In order to approximate a real hand functionality, a myoelectric prosthesis should implement a large number of hand and finger gestures. However, increasing number of gestures can lead to a decrease in recognition accuracy. In this work a Myo arm band device is used to recognize fourteen gestures (five build in gestures of Myo armband in addition to nine new gestures). The data in this research is collected from three body-able subjects for a period of 7 seconds per gesture. The proposed method uses a pattern recognition technique based on Multi-Layer Perceptron Neural Network (MLPNN). The results show an average accuracy of 90.5% in recognizing the proposed fourteen gestures.

Sign language is different from spoken language that emphasizes both the audio and verbal aspects. There are two sign languages applicable in Indonesia, namely Indonesian Signal System (SIBI) and Indonesian Sign Language (BISINDO). SIBI converts spoken Indonesian language into sign language and follows the Indonesian spoken language’s grammatical structure complete with prefix and suffix. In contrast to SIBI, BISINDO translates one word from the Indonesian spoken language in accordance with its context followed by an expression showing the ongoing events. We choose BISINDO rather than SIBI in line with the deaf’ people’s suggestions and requests out there to make BISINDO as an official Indonesian sign language to replace SIBI. This research aims to develop a translator model of BISINDO through computer vision technology such as Microsoft Kinect XBox and machine translation using Hidden-Markov Model (HMM) with optimal number of hidden states. We utilize skeleton data from Kinect sensor for feature extraction. There are four kinds of skeleton features used in this study consisting of the movement of the shoulders, upper arms, forearms, and hands. The experiment results by using this methodology successfully recognize the gesture of BISINDO with an accuracy is around 60%.

Sign language is the language used by the deaf to communicate. In general, there are two types of sign languages used in Indonesia, namely SIBI (Indonesian sign language system) and BISINDO (Indonesian sign language). Of the two, the sign language most often used by the deaf is BISINDO, because BISINDO is considered easier to understand. This study aims to explain the use of BISINDO in Deaf Indonesian language classes in the Surakarta Gerkatin organization. The method used in this study is descriptive qualitative with data in the form of image documentation. The data source is in the form of sign language learning process. Data collection techniques are carried out through observation, recording and note-taking techniques. Data analysis techniques use interactive models. The results showed that the movements used in BISINDO learning at Gerkatin Surakarta involved the use of facial expressions, body language, arms and hands. There is a lot of vocabulary taught such as learning words, Indonesian, not yet, like, where, like, small, eat, alone, understand, don't know, work, sleep, words, healthy, black, fat, thin, color, about, purpose, lesson, early. If you look closely at the vocabulary taught, it has different conceptual sign languages. Like a like sign, demonstrated with the position of the hands in front of the mouth, then the palms open up plus the movement of the hands sticking from the chest towards the other person. Small gesture, in front of the chest with the tip of the thumb against the ring finger and the other three fingers open upwards, accompanied by a plucking motion of the two attached fingers. All the movements taught are made natural so that they are easy to understand according to the interpretation of the deaf. Thus, it is hoped that this variation of BISINDO's sign language will be increasingly recognized and widely used in teaching deaf sign language.

Sign Language Recognition (SLR) has been an active area of research, but sentence-level SLR remains relatively underexplored. While most studies focus on recognizing individual signs, understanding full sentences presents greater challenges. This research proposes a sentence-level SLR using a combination of 3D Convolutional Neural Networks (3D CNN) for spatio-temporal feature extraction with sequential modeling using Long Short-Term Memory (LSTM) and Bidirectional LSTM (BiLSTM). Connectionist Temporal Classification (CTC) is also used to enable training without word-level annotations. In this study, we used the Indonesian Sign Language (BISINDO) dataset, specifically the DKI Jakarta version, consisting of 900 videos representing 30 sentences, which was expanded to 3600 videos through data augmentation techniques such as speed variation and brightness adjustments. All videos underwent preprocessing to ensure data quality, and Bayesian Optimization was applied for hyperparameter tuning to obtain optimal configurations for each model. Both models were trained with CTC loss and evaluated using Word Error Rate (WER). The 3DCNN-LSTM model achieved a WER result of 59.21%, while the 3DCNN-BiLSTM presents a significantly better performance with a WER of 2.77%. Despite these promising results, the models’ ability to generalize across different signers may require further research, as the dataset used in this research involved only a single signer.

Code examples are important resources for expressing correct application programming interface (API) usages. However, many framework and library APIs fail in offering sufficient code examples in corresponding API documentations. This is because constructing proper code examples for documentation purpose takes significant developers' efforts. To reduce such effort, this work proposes a methodology, PropER-Doc, that recommends proper code examples for documentation purpose. PropER-Doc accepts queries from API developers and utilizes code search engines (CSEs) to collect corresponding code example candidates. The structural and conceptual links between API elements are captured from the API implementation and available API documents to guide candidate recommendation. During recommendation, PropER-Doc groups collected candidates based on involved API types for distinguishing different API usages. To assist API developers in selecting proper candidates, a diagrammatic presentation and three code example appropriateness metrics are also developed in {PropER-Doc}. Two case studies on Eclipse JDT framework are conducted to confirm the effectiveness of PropER-Doc.

Application Programming Interfaces (APIs) are everyday tools for every software professional. When creating an API, a designer typically abstracts the tasks the API intends to perform. The API user, on the other hand, should use the API to perform specific-propose tasks. During the development of an API, the designer needs to write dialogues with which the user will interact with the API, performing a conversation between them. In this way, an API capable of offering effective dialogues to its users is called a conversational API. A conversational API is the one that is capable of communicating to its users its form of use and its internal logic of operation, making it clear the design decisions when abstracting concepts and tasks. In this paper, we present Colloquy, an evidence-based method to support the design of conversational APIs. Colloquy was conceived based on the lessons learned from an action research we conducted to identify effective strategies for designing conversational APIs. For six months, we followed the redesign of an existing API developed at an R&D laboratory of a large IT company. In this paper, we also present the first empirical evaluation of Colloquy. We conducted a case study in which the method was employed to design an API for refactoring source code in the Java programming language. Among other benefits, we found that Colloquy was effective in developing empathy with users and modeling conversations with them. Besides, Colloquy contributed to identifying new requirements and created models as API documentation.

APIs (application programming interfaces) play a key role in software development. Virtually, all programmers are potential users of third-party APIs. From the perspective of the theory of Semiotic Engineering, we may characterize an API as an artifact mediating the communication between two types of developers: the API designers and its users. During the construction of an API, the designers should establish proper dialogues with the users. These dialogues will enable the conversation of these actors at the interaction time. In this way, we define a conversational API as an API capable of offering effective dialogues to its users. In this paper, we introduce Colloquy, a method for supporting the design of conversational APIs. Colloquy results from the lessons learnt during an action research conducted for redesigning a real and complex API. The set of Colloquy resources allow API designers to go beyond conventional concerns with usability. We also report in this paper a case study in which Colloquy was used for redesigning a refactoring API. The study findings indicate the method helped the designer creating empathy with the API users, as well as better reflecting and depicting the requirements and the conversations that the API should attend to the different user profiles.

Indonesian Sign Language (BISINDO) is an essential means of communication for deaf people in Indonesia. However, communication challenges persist for the general public, who lack sign language proficiency. To bridge this gap, a real-time BISINDO recognition system is essential. This research compares the performance of several distance models in recognizing BISINDO letters using a camera-based recognition system. The study evaluated three distance methods City Block, Canberra, and Chebyshev applied to shape feature extraction through chain code analysis. Testing was conducted under both indoor and outdoor conditions to assess system robustness. The results revealed that the Chebyshev distance method achieved the highest accuracy in indoor scenarios, particularly when participants wore white clothing, reaching an accuracy of 100%. Meanwhile, in outdoor conditions, the Canberra distance method outperformed the others, achieving the highest accuracy of 96.15% with white clothing. However, outdoor accuracy was generally lower due to uncontrollable lighting, which affected recognition. Despite these challenges, the system demonstrated its capability to recognize BISINDO letters effectively in both environments. This research highlights the importance of choosing the appropriate distance model and controlling external factors, such as lighting, to improve system performance. The findings are expected to significantly advance efficient and reliable sign language recognition systems in Indonesia, thereby enhancing communication between the deaf community and the broader population. This development is a step forward in promoting inclusivity and accessibility in everyday interactions.

Indonesian Sign Language or BISINDO is an alternative language used by people who suffer from disabilities, especially those who have hearing impairments. This language grew and developed from the deaf community, so its use is based on the visual aspect. This research aims to apply Residual Networks to detect objects in the context of Bisindo Letter Sign Language, with the hope of increasing accuracy and efficiency in letter recognition. Object detection goes through 2 stages, namely feature extraction and model training. ResNet is a type of Convolutional Neural Network (CNN) architecture that utilizes models that have been previously trained, so it can save the time required in the model development process. In this research, Residual Network (ResNet) was used for feature extraction to recognize important aspects in the Bisindo letter sign image, such as hand position, finger shape characteristics, and direction of movement. The research results show that the new dataset used as training data and test data has a fairly good ability to detect with a division of 70% train set, 20% valid set and 10% test set with size 640x640 with 300 epochs for the training model.

This research develops a system for recognizing finger spelling gestures in Indonesian Sign Language (BISINDO) using Convolutional Neural Network (CNN). The objective of this research is to apply the Convolutional Neural Network (CNN) method to the BISINDO finger spelling gesture recognition system to improve its accuracy. The method employed is Convolutional Neural Network (CNN), an effective method for processing image data for pattern recognition. Based on the test results, the system demonstrates that the developed CNN model is capable of recognizing BISINDO finger spelling gestures with an accuracy of 97.5%. This indicates that the BISINDO finger spelling gesture recognition system performs well in pattern recognition. The implementation of the system for real-time prediction via a web interface using Flask also enhances its accessibility. However, there is still room for improvement, particularly in recognizing one of the 26 letters that has not been predicted accurately. For further development, it is recommended to consider collecting a larger dataset and incorporating more complex gesture variations to improve recognition accuracy.

Variation in Indonesian Sign Language: A typological and sociolinguistic analysis

7 Accounting for variation in Indonesian Sign Language

External service integration and adherence to industry standards has become ever more important for collections data management platforms. External APIs (Application Programming Interfaces), allow for the development of bi-directional data flows critical to service integration. In contrast to service-oriented backend APIs, public APIs must have continually up-to-date, comprehensive documentation that covers common use cases, on-the-fly request validation, and meaningful error messages. OpenAPI (OpenAPI Initiative 2021), a machine-readable API documentation specification can help significantly with testing and maintenance, and libraries can be used to automate common maintenance tasks. Specify 7 is a biological collections data management platform developed by the Specify Collections Consortium (Specify Software Consortium 2021). This presentation summarizes the challenges and lessons learned with publishing the existing backend Specify 7 API to a public-facing external API. Each Specify 7 API is composed of 200 resources. A standard set of CRUD (Create, Read, Update, Delete) operations is provided for each resource for client interaction with a group of service-based endpoints for bulk operations such as file uploads, file-based data imports, and attachment manipulation. To support the migration, we developed a custom library to enhance request validation. Parameter validation is extended through a real-time comparison against the existing schema and data. The library is available to the community under a MIT license on GitHub (https://github.com/specify/open_api_tools/). In this presentation, we will close with an overview of the next steps for the Specify 7 public API. These include: An update to the latest OpenAPI specification, version 3.1. The latest version aims to increase compatibility with the Javascript Object Notation (JSON) Schema specification, and thus would allow us to use JSON Schema (IETF Trust 2021) validation frameworks. An in-depth evaluation of GraphQL for its ability to force all endpoints to be strongly typed and automatic validation of request parameters and response objects. An update to the latest OpenAPI specification, version 3.1. The latest version aims to increase compatibility with the Javascript Object Notation (JSON) Schema specification, and thus would allow us to use JSON Schema (IETF Trust 2021) validation frameworks. An in-depth evaluation of GraphQL for its ability to force all endpoints to be strongly typed and automatic validation of request parameters and response objects.

CORDITE: The Curated CORona Drug InTERactions Database for SARS-CoV-2