A Survey on Low-Latency DNN-Based Speech Enhancement
Sensors · 26.01.2023 · 23 (3) · 1380
DOI: 10.3390/s23031380 ↗Kurz gefasst
Übersichtsarbeit in der Zeitschrift Sensors, online erschienen am 26.01.2023. Verfasst von Szymon Drgas. Beteiligte Einrichtung: Institute of Automatic Control and Robotics, Poznan University of Technology, Piotrowo 3A Street, 60-965 Poznan, Poland. Hauptthemen laut MEDLINE: Speech, Neural Networks, Computer. Die Arbeit wurde bisher 26-mal zitiert (Crossref) und nennt 117 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: subvention financial means.
Abstract (Originalsprache)
This paper presents recent advances in low-latency, single-channel, deep neural network-based speech enhancement systems. The sources of latency and their acceptable values in different applications are described. This is followed by an analysis of the constraints imposed on neural network architectures. Specifically, the causal units used in deep neural networks are presented and discussed in the context of their properties, such as the number of parameters, the receptive field, and computational complexity. This is followed by a discussion of techniques used to reduce the computational complexity and memory requirements of the neural networks used in this task. Finally, the techniques used by the winners of the latest speech enhancement challenges (DNS, Clarity) are shown and compared.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Institute of Automatic Control and Robotics, Poznan University of Technology, Piotrowo 3A Street, 60-965 Poznan, Poland.
ORCID-Profile: Szymon Drgas
Förderung
- subvention financial means
Zitieren
APA
Drgas, S. (2023). A Survey on Low-Latency DNN-Based Speech Enhancement. Sensors, 23(3), 1380. https://doi.org/10.3390/s23031380
Vancouver
Drgas S. A Survey on Low-Latency DNN-Based Speech Enhancement. Sensors. 2023;23(3):1380. doi:10.3390/s23031380
BibTeX
@article{Drgas2023,
author = {Drgas, Szymon},
title = {A Survey on Low-Latency DNN-Based Speech Enhancement},
journal = {Sensors},
year = {2023},
volume = {23},
number = {3},
pages = {1380},
doi = {10.3390/s23031380}
}
EUVIMED