A scalable codec for bone-conducted speech based on generative and diffusion models
The Journal of the Acoustical Society of America · 2025
DOI: 10.1121/10.0038967 ↗Kurz gefasst
Fachartikel in der Zeitschrift The Journal of the Acoustical Society of America, 2025. Verfasst von Xiaolong Hu und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Information and Communication Engineering, Dalian University of Technology (DUT), Dalian, 116023, China. Hauptthemen laut MEDLINE: Bone Conduction, Speech Acoustics, Signal Processing, Computer-Assisted, Speech. Die Arbeit nennt 64 Literaturangaben. Gefördert von: Natural Science Foundation of China, National High Technology Research and Development Programof China, Natural Science Foundation of Liaoning Province of China.
Abstract (Originalsprache)
In extremely noisy communication scenarios, the bone-conducted microphone (BCM) speech codec is often combined with speech bandwidth extension to improve the BCM speech quality. However, this tandem approach leads to a complex system architecture. To address the problem, a scalable codec for BCM speech based on generative and diffusion probabilistic models is proposed in this paper. Specifically, a specialized codec architecture is constructed to encode BCM speech while complementing its high-frequency components. Then, a key feature extraction block is presented to address the diminishing memory capacity in shallow layers as the network depth increases. Next, considering the potential lack of high-frequency detail information, an overall refinement block is introduced to refine the reconstructed speech signals. Finally, based on the U-Net architecture, a diffusion probability model is proposed to upsample the input audio signal from a bandwidth of 8 kHz to a high-resolution audio signal with a bandwidth of 20 kHz and a sampling rate of 48 kHz. The proposed method can simultaneously encode and improve BCM speech quality using a single network. It supports different bitrate settings without architectural changes or retraining and dynamically adjusts transmitted data based on changing network load. Simulation experiments demonstrate its feasibility.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Department of Information and Communication Engineering, Dalian University of Technology (DUT), Dalian, 116023, China.
ORCID-Profile: Xiaolong Hu, Zhe Chen, Fuliang Yin
Förderung
- Natural Science Foundation of China
- National High Technology Research and Development Programof China
- Natural Science Foundation of Liaoning Province of China
- Fundamental Research Fund for the Central Universities of China
Zitieren
APA
Hu, X., Chen, Z., & Yin, F. (2025). A scalable codec for bone-conducted speech based on generative and diffusion models. The Journal of the Acoustical Society of America. https://doi.org/10.1121/10.0038967
Vancouver
Hu X, Chen Z, Yin F. A scalable codec for bone-conducted speech based on generative and diffusion models. The Journal of the Acoustical Society of America. 2025. doi:10.1121/10.0038967
BibTeX
@article{Hu2025,
author = {Hu, Xiaolong and Chen, Zhe and Yin, Fuliang},
title = {A scalable codec for bone-conducted speech based on generative and diffusion models},
journal = {The Journal of the Acoustical Society of America},
year = {2025},
doi = {10.1121/10.0038967}
}
EUVIMED