Hand Sign to Bangla Speech: A Deep Learning in Vision Based System for Recognizing Hand Sign Digits and Generating Bangla Speech

Ahmed, Shahjalal; Islam, Md. Rafiqul; Hassan, Jahid; Ahmed, Minhaz Uddin; Ferdosi, Bilkis Jamal; Saha, Sanjay; Shopon,

doi:10.2139/ssrn.3358187

Cited by 20 publications

(13 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, Hu et al had developed a suggestion for the design of hybrid CNN and RNN to preserve the temporal features correctly for the electromyogram signal, which addresses the issue of action identification. Another work [26] describes an extraordinary CNN model that automatically detects numbers relying on hand signals and communicates the specific outcome in the Bangla language, which is followed in this study. In a similar work [27], a CRNN module for hand pose estimation is conducted.…”

Section: Related Workmentioning

confidence: 99%

Sign Language Recognition for Arabic Alphabets Using Transfer Learning Technique

Zakariah

Alotaibi

Koundal

et al. 2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

Sign language is essential for deaf and mute people to communicate with normal people and themselves. As ordinary people tend to ignore the importance of sign language, which is the mere source of communication for the deaf and the mute communities. These people are facing significant downfalls in their lives because of these disabilities or impairments leading to unemployment, severe depression, and several other symptoms. One of the services they are using for communication is the sign language interpreters. But hiring these interpreters is very costly, and therefore, a cheap solution is required for resolving this issue. Therefore, a system has been developed that will use the visual hand dataset based on an Arabic Sign Language and interpret this visual data in textual information. The dataset used consists of 54049 images of Arabic sign language alphabets consisting of 1500\ images per class, and each class represents a different meaning by its hand gesture or sign. Various preprocessing and data augmentation techniques have been applied to the images. The experiments have been performed using various pretrained models on the given dataset. Most of them performed pretty normally and in the final stage, the EfficientNetB4 model has been considered the best fit for the case. Considering the complexity of the dataset, models other than EfficientNetB4 do not perform well due to their lightweight architecture. EfficientNetB4 is a heavy-weight architecture that possesses more complexities comparatively. The best model is exposed with a training accuracy of 98 percent and a testing accuracy of 95 percent.

show abstract

Section: Related Workmentioning

confidence: 99%

Sign Language Recognition for Arabic Alphabets Using Transfer Learning Technique

Zakariah

Alotaibi

Koundal

et al. 2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

show abstract

“…We implemented rotation testing cases with the systems Fig. 7: Comparative analysis of each system's accuracy for rotated images [5], [14]- [17] for testing their system. From the graph we identified that system [17] works poorly for rotated images.…”

Section: Test Case-1 Rotationmentioning

confidence: 99%

“…If we got 100 converted images from the input, then the concept is clear getting lots of images from a sample image. From the graph we identified that system [17] works poorly for contrasted images.…”

Section: Test Case-2 Contrastmentioning

confidence: 99%

“…In Fig. 3 has shown the background images use as a testing case to test the existing system [5], [17]. When an image input in the proposed framework then converted the image with different background.…”

Section: F Test Case-4 Backgroundmentioning

confidence: 99%

“…In the graph x-axis showing the number of sample and the yaxis showing the existing system accuracy according to the testing cases. We implemented different background testing cases with the systems [5], [14]- [17] for testing their system. From the graph we identified that system [14] works poorly for different background images.…”

Section: F Test Case-4 Backgroundmentioning

confidence: 99%

See 2 more Smart Citations

An Automated Testing Framework for Gesture Recognition System using Dynamic Image Pattern Generation with Augmentation

Ashaduzzaman

Hasan

Islam

et al. 2021

GUB JOURNAL OF SCIENCE AND ENGINEERING

View full text Add to dashboard Cite

In the field of information technology, the gesture recognition system plays a very essential role. As it has achieved vast importance, it is mandatory to test the recognition system to ensure the quality of the system by identifying the bugs in the software. In our research, we suggested a dynamic testing method for gesture recognition software. using dynamic image pattern generation with augmentation. The automated software testing framework is a set of processes to create new test cases for properly testing a image processing software. The research intention for generate automated testing cases by following a standard process which helps to increase the performance and efficiency of the gesture recognition system. We have built the framework to give proper testing and give result (accuracy and defect) for which gesture recognition system already in the market. our research, the team strongly following and adding two software testing standard. First one is ISO/IEC/IEEE/291129-3 to define the process for testing software. And the second one is ISO/IEC/IEEE/291129-5 to implement the techniques for software testing. We proposed this framework with major five parameters by noise, rotation, background, contrast, and scale. Which are the most use with every gesture recognition system. Our developed framework’s phase is used to generate new testing cases based on the existing gesture recognition system’s data. There are we work with five systems, commonly with the gesture recognition for experiments. We provide the testing report with total accuracy and defect by comparing existing well-known system’s data. At the final result, our system suggested an analysis report based on the testing result. And tell what are the improvement needs for the existing system to consider noised images or different scaled images to build a robust system. GUB JOURNAL OF SCIENCE AND ENGINEERING, Vol 7, Dec 2020 P 42-50

show abstract

A review on computational methods based automated sign language recognition system for hearing and speech impaired community

Robert

Hemanth

2023

Concurrency and Computation

View full text Add to dashboard Cite

Summary The recent advancements in computer vision and deep learning have led to promising progress in various motion detection and gesture recognition methods. Thriving efforts in the field of sign language recognition (SLR) during recent years led to interaction between humans and computer systems. Contributing a real‐time automated Sign Language recognition system will be a remarkable treasure for hearing and speech‐impaired people that will break the barriers of interaction with the real world. Albeit several research works are accomplished in sign language recognition, there is still a demand for developing a real‐time automated sign language recognition system. Compared to other methodologies, the techniques adopted may have advantages and disadvantages, and they vary from researcher to researcher. There are still some issues with employing these SLR models and procedures regularly, even though numerous research studies have been conducted to determine the most acceptable methods and models for sign language recognition. It gets more expensive and difficult in terms of resources when the developed automated SLR system becomes available as a product. Hence, for the welfare of the hearing and speech‐impaired community, the researchers are still endeavoring way to find a cost‐effective method. This work brings forth the challenges faced by scientists in developing a cost‐effective commercial prototype for the hearing and speech‐impaired community. This paper also explores and analyses various deep learning techniques and methods used in developing a sign language recognition system. The objective behind this work is to identify the best method which produces high accuracy in developing a cost‐effective sign language recognition system aiding communication between signers and non‐signers so that they can be a part of growing technology.

show abstract

Hand Sign to Bangla Speech: A Deep Learning in Vision Based System for Recognizing Hand Sign Digits and Generating Bangla Speech

Cited by 20 publications

References 7 publications

Sign Language Recognition for Arabic Alphabets Using Transfer Learning Technique

Sign Language Recognition for Arabic Alphabets Using Transfer Learning Technique

An Automated Testing Framework for Gesture Recognition System using Dynamic Image Pattern Generation with Augmentation

A review on computational methods based automated sign language recognition system for hearing and speech impaired community

Contact Info

Product

Resources

About