Efficient Speech Detection in Environmental Audio Using Acoustic Recognition and Knowledge Distillation
Sensors · 2024
DOI: 10.3390/s24072046 ↗Kurz gefasst
Fachartikel in der Zeitschrift Sensors, 2024. Verfasst von Drew Priebe und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Cognitive Science and Artificial Intelligence, Tilburg University, 5037 Tilburg, The Netherlands. Hauptthemen laut MEDLINE: Speech, Voice. Die Arbeit wurde bisher 12-mal zitiert (Crossref) und nennt 40 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Mambo, Guarden.
Abstract (Originalsprache)
The ongoing biodiversity crisis, driven by factors such as land-use change and global warming, emphasizes the need for effective ecological monitoring methods. Acoustic monitoring of biodiversity has emerged as an important monitoring tool. Detecting human voices in soundscape monitoring projects is useful both for analyzing human disturbance and for privacy filtering. Despite significant strides in deep learning in recent years, the deployment of large neural networks on compact devices poses challenges due to memory and latency constraints. Our approach focuses on leveraging knowledge distillation techniques to design efficient, lightweight student models for speech detection in bioacoustics. In particular, we employed the MobileNetV3-Small-Pi model to create compact yet effective student architectures to compare against the larger EcoVAD teacher model, a well-regarded voice detection architecture in eco-acoustic monitoring. The comparative analysis included examining various configurations of the MobileNetV3-Small-Pi-derived student models to identify optimal performance. Additionally, a thorough evaluation of different distillation techniques was conducted to ascertain the most effective method for model selection. Our findings revealed that the distilled models exhibited comparable performance to the EcoVAD teacher model, indicating a promising approach to overcoming computational barriers for real-time ecological monitoring.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Cognitive Science and Artificial Intelligence, Tilburg University, 5037 Tilburg, The Netherlands.
- Naturalis Biodiversity Center, 2333 Leiden, The Netherlands.
ORCID-Profile: Burooj Ghani, Dan Stowell
Förderung
- Mambo
- Guarden
Zitieren
APA
Priebe, D., Ghani, B., & Stowell, D. (2024). Efficient Speech Detection in Environmental Audio Using Acoustic Recognition and Knowledge Distillation. Sensors. https://doi.org/10.3390/s24072046
Vancouver
Priebe D, Ghani B, Stowell D. Efficient Speech Detection in Environmental Audio Using Acoustic Recognition and Knowledge Distillation. Sensors. 2024. doi:10.3390/s24072046
BibTeX
@article{Priebe2024,
author = {Priebe, Drew and Ghani, Burooj and Stowell, Dan},
title = {Efficient Speech Detection in Environmental Audio Using Acoustic Recognition and Knowledge Distillation},
journal = {Sensors},
year = {2024},
doi = {10.3390/s24072046}
}
EUVIMED