MULTI-STAGE TRAINING FOR CROSS-DOMAIN FULL-BAND AUDIO PACKET LOSS CONCEALMENT

被引：0

作者：

Li, Nan ^{[1
]}

Yu, Guochen ^{[1
]}

Zhang, Chen ^{[1
]}

Zhou, Chao ^{[1
]}

Huang, Qi ^{[1
]}

Yu, Bing ^{[1
]}

机构：

[1] Kuaishou Technol, Beijing, Peoples R China

来源：

2024 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING WORKSHOPS, ICASSPW 2024 | 2024年

关键词：

packet loss concealment; multi-stage training; cross-domain;

D O I：

10.1109/ICASSPW62465.2024.10626444

中图分类号：

O42 [声学];

学科分类号：

070206 ; 082403 ;

摘要：

This paper introduces a multi-stage full-band packet loss concealment (PLC) method that incorporates both time-domain and frequency-domain learning. Specifically, the time-domain model is elaborately designed to generate the missing packets, while the frequency-domain model aims to remove discontinuity and noise. The proposed system achieved the 1st place ranking in the ICASSP 2024 PLC Challenge, demonstrating superior performance with a mere 20ms algorithm latency.

引用

页码：35 / 36

页数：2

共 24 条

[21] Multi-modal cross-domain self-supervised pre-training for fMRI and EEG fusion
Wei, Xinxu
Zhao, Kanhao
Jiao, Yong
Carlisle, Nancy B.
Xie, Hua
Fonzo, Gregory A.
Zhang, Yu
NEURAL NETWORKS, 2025, 184
[22] Cross-Domain Learning Framework for Tracking Users in RIS-Aided Multi-Band ISAC Systems With Sparse Labeled Data
Hu, Jingzhi
Niyato, Dusit
Luo, Jun
IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS, 2024, 42 (10) : 2754 - 2768
[23] A two-stage importance-aware subgraph convolutional network based on multi-source sensors for cross-domain fault diagnosis
Yu, Yue
He, Youqian
Karimi, Hamid Reza
Gelman, Len
Cetin, Ahmet Enis
NEURAL NETWORKS, 2024, 179
[24] MF-GSLAE: A Multi-Factor User Representation Pre-Training Framework for Dual-Target Cross-Domain Recommendation
Wang, Hao
Yin, Mingjia
Zhang, Luankang
Zhao, Sirui
Chen, Enhong
ACM TRANSACTIONS ON INFORMATION SYSTEMS, 2025, 43 (02)

← 1 2 3 →