Loading article...
Loading article...
Stream/Area:
-
Subject:
-
Paper ID:
IJSREM63783
Volume:
10 |
5 |
May 2026
Pages:
1-4
Publication Frequency:
Monthly (12 issues per year)
License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.
MASTER OF COMPUTER APPLICATION
BAPUJI INSTITUTE OF ENGINEERING AND TECHNOLOGY
MASTER OF COMPUTER APPLICATION
BAPUJI INSTITUTE OF ENGINEERING AND TECHNOLOGY
Interview evaluation plays a crucial role in academic admissions and recruitment processes. Traditional evaluation methods are subjective, time-consuming, and prone to inconsistency due to human bias and variability in judgment. This paper presents EVALORA, an AI-driven automated interview evaluation platform designed to standardize, accelerate, and enhance the assessment process. The proposed system processes recorded interview videos by extracting audio, performing noise reduction and normalization, converting speech to text using advanced Automatic Speech Recognition (ASR) models, and analyzing responses using Large Language Models (LLMs). The system automatically identifies question–answer pairs and evaluates responses based on relevance, technical accuracy, fluency, and depth of knowledge. EVALORA integrates OpenAI Whisper for speech-to-text conversion and GPT-based language models for contextual evaluation. A Flask-based backend and web-based interface ensure usability and scalability. Experimental validation demonstrates reliable performance across functional and non-functional requirements, including usability, performance, reliability, and security. The system reduces evaluation time, improves consistency, and provides structured qualitative feedback, making it suitable for campus placements, recruitment screening, and academic assessments.
KEYWORDS
We had developed Evalora, an AI-enabled interview evaluation system to automate and standardize the way of evaluating interview performance. Modern AI methods are applied to mitigate the limitations of conventional interview evaluation tools based on subjectivity, inconsistency and the amount of time required. Evalora uses an organised and indication, modular pipeline to analyse recorded interviewer interviewee videos via video-to-audio extraction, pre-processing and AI-based audio answer evaluation. Through reviewing responses using predefined criteria including relevancy, correctness, fluency and technical depth, the system delivers standardized and impartial assessment results. The pre-trained AI models allow for the efficient realisation with no massive data collection or model training. The system was evaluated in various functional and non-functional areas of reliability, usability and performance. The test results show that the system is functioning and generating meaningful evaluation reports. The user-friendly web interface offers an easy-to-understand result display for good usability. In conclusion, Evalora proves the real-life usage of Artificial Intelligent for interview assessment. The initiative shows how AI-enabled systems can help humans make more consistent and efficient decisions. By adding emotion analysis, support of more languages and real-time corrections the current prototype could grow into a fully-fledged interview evaluation system.
1. OpenAI, Whisper: Robust Speech Recognition via Strong Large-Scale Unsupervised Data Augmentation, OpenAI Documentation, 2022. 2. OpenAI, GPT Models and API Documentation -, OpenAI Platform, https://platform.openai.com/docs 3. Unpacking the contents of a Flask Documentation, Flask Web Framework, https://flask.palletsprojects.com/ 4. McKinney, W., Python for Data Analysis, O’Reilly Media, 2017. 5. Pedregosa, F. et al., “Scikit-learn: Machine Learning in Python,” Journal of Machine Learning Research, Vol. 12, 2011. 6. Librosa Development Team, The Librosa: Python Library for Audio and Music Analysis, https://librosa.org/
Publication Date
5/22/2026
Peer Review Type
Double-Blind Peer Review
Review Rounds: 2
Documents
Article Views
0
Total views since publication
PDF Downloads
2
Total downloads since publication
Geographic Reach
5
Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.
Plagiarism Screening
This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.
Conflict of Interest
The authors declare no conflict of interest related to this research work.
Funding Information
No funding information available for this research.
Related research in the same subject area
Loading similar articles...