Using Whisper Embeddings for Audio-Only Latent Token Classification of Classroom Management Practices

Wesley Griffith Morris, Jessica Vitale, Isabel Arvelo


Abstract
In this study, we developed a textless NLP system using a fine-tuned Whisper encoder to identify classroom management practices from noisy classroom recordings. The model segments teacher speech from non-teacher speech and performs multi-label classification of classroom practices, achieving acceptable accuracy without requiring transcript generation.
Anthology ID:
2025.aimecon-main.17
Volume:
Proceedings of the Artificial Intelligence in Measurement and Education Conference (AIME-Con): Full Papers
Month:
October
Year:
2025
Address:
Wyndham Grand Pittsburgh, Downtown, Pittsburgh, Pennsylvania, United States
Editors:
Joshua Wilson, Christopher Ormerod, Magdalen Beiting Parrish
Venue:
AIME-Con
SIG:
Publisher:
National Council on Measurement in Education (NCME)
Note:
Pages:
154–162
Language:
URL:
https://aclanthology.org/2025.aimecon-main.17/
DOI:
Bibkey:
Cite (ACL):
Wesley Griffith Morris, Jessica Vitale, and Isabel Arvelo. 2025. Using Whisper Embeddings for Audio-Only Latent Token Classification of Classroom Management Practices. In Proceedings of the Artificial Intelligence in Measurement and Education Conference (AIME-Con): Full Papers, pages 154–162, Wyndham Grand Pittsburgh, Downtown, Pittsburgh, Pennsylvania, United States. National Council on Measurement in Education (NCME).
Cite (Informal):
Using Whisper Embeddings for Audio-Only Latent Token Classification of Classroom Management Practices (Morris et al., AIME-Con 2025)
Copy Citation:
PDF:
https://aclanthology.org/2025.aimecon-main.17.pdf