Tri-level Speech-Based Student Attendance Monitoring using LSTM

This paper introduces an advanced attendance tracking system utilizing Automatic Speech Recognition (ASR) technology based on Long Short-Term Memory (LSTM) networks. Through the integration of Mel Frequency Cepstral Coefficients (MFCC), Short-Time Fourier Transform (STFT), and wavelet analysis, the system achieves exceptional accuracy in digit recognition. Each student is uniquely identified by their assigned roll number, represented by digits. When a student vocalizes their respective roll number, the system records their attendance. Additionally, the system employs three-level verification methods to ensure enhanced authentication. Through the integration of STFT and wavelet transforms, the system effectively addresses the limitations of each technique, resulting in improved recognition accuracy and operational efficiency. Overall, this work endeavors to modernize attendance tracking by introducing innovative features for robust authentication.