Date of Award
11-30-2022
Document Type
Thesis
Degree Name
Master of Science (MS)
Department
Computer Science
First Advisor
Mariofanna Milanova
Abstract
For both teachers and students studying science, technology, engineering, and mathematics (STEM), active learning is more likely to be productive because learners engage in a variety of classroom activities. As instructors are trying different pedagogies in classroom, it is also important to check the effectiveness of those methods. The aim of our work is to identify the classroom activities with more accuracy which will help to measure the student involvement in the class. Using automatic audio classification, we can help to improve active learning strategies in the classroom, and it will be cost effective too. Various deep learning techniques, such as deep neural networks, convolutional neural networks and long short-term (LSTM) memories, are examined in this study for categorizing classroom audio. We test the models’ using recordings from our classroom for three different types of tasks labeled "single voice," "multiple voice," and "no voice." To train the model, the audio recording's generated Mel spectrogram is employed. We get the highest accuracy of 98% and F1 score of .97 with the LSTM with 10 sec frames.
Recommended Citation
Mou, Afsana Rahman, "Classroom Audio Classification Using Deep Learning Frameworks" (2022). Theses and Dissertations. 1102.
https://research.ualr.edu/etd/1102
