Authors : Tanishq Jaiswal; Varsha Teeratipally; Ritendu Bhattacharyya; Bharani Kumar Depuru

Volume/Issue : Volume 9 - 2024, Issue 1 - January

Google Scholar : http://tinyurl.com/yu256hkt

Scribd : http://tinyurl.com/fj8acszu

DOI : https://doi.org/10.5281/zenodo.10634653

Abstract : Ai-based assessment scrutiny is the most convenient and precise method to eliminate the repetitive task of answer grading; consisting of text extraction methodologies and using Deep Learning Architecture to evaluate with reference to the correct answer and Question provided. In the landscape of educational assessment, the traditional methods of answer evaluation face challenges in adapting to the dynamic and evolving nature of learning. This paper proposes a complete end- to-end answer-grading architecture that can be deployed to provide an interface for a fully automated- Deep- learning answer-grading mechanism. This research introduces a groundbreaking approach to address these challenges, presenting a solution that seamlessly integrates advanced text extraction and deep learning architectures. Our objective is to achieve unparalleled precision in answer evaluation, setting a new standard in the field. Our method involves the extraction of audio files, precise text extraction from audio, and a Deep Neural Networks DNN-based model for answer evaluation, based on a database that provides the correct answer and relevant data is fetched. Proposing a reliable, accurate, easy-to- deploy best-in-class technology to eradicate manual repetitive tasks. Providing a very user-friendly interface to the student, and a dynamic backend to monitor results along with the high level of precision. These AI-based evaluation methods can be used in numerous places in the evolving Education industry providing students with a convenient interface and automation. The objective is to elevate the precision and adaptability of answer assessment methodologies in the dynamic landscape of modern education. The educational landscape continues to evolve, our research not only addresses current challenges but also lays the groundwork for future advancements in the field of educational assessment, promising a new era of precision and adaptability. This paper includes text extraction from architecture-based Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), and transformers like an encoder-decoder transformer (whisper).

Keywords : Ai-based assessment scrutiny is the most convenient and precise method to eliminate the repetitive task of answer grading; consisting of text extraction methodologies and using Deep Learning Architecture to evaluate with reference to the correct answer and Question provided. In the landscape of educational assessment, the traditional methods of answer evaluation face challenges in adapting to the dynamic and evolving nature of learning. This paper proposes a complete end- to-end answer-grading architecture that can be deployed to provide an interface for a fully automated- Deep- learning answer-grading mechanism. This research introduces a groundbreaking approach to address these challenges, presenting a solution that seamlessly integrates advanced text extraction and deep learning architectures. Our objective is to achieve unparalleled precision in answer evaluation, setting a new standard in the field. Our method involves the extraction of audio files, precise text extraction from audio, and a Deep Neural Networks DNN-based model for answer evaluation, based on a database that provides the correct answer and relevant data is fetched. Proposing a reliable, accurate, easy-to- deploy best-in-class technology to eradicate manual repetitive tasks. Providing a very user-friendly interface to the student, and a dynamic backend to monitor results along with the high level of precision. These AI-based evaluation methods can be used in numerous places in the evolving Education industry providing students with a convenient interface and automation. The objective is to elevate the precision and adaptability of answer assessment methodologies in the dynamic landscape of modern education. The educational landscape continues to evolve, our research not only addresses current challenges but also lays the groundwork for future advancements in the field of educational assessment, promising a new era of precision and adaptability. This paper includes text extraction from architecture-based Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), and transformers like an encoder-decoder transformer (whisper).