Federated Learning based Audio Semantic Communication over Wireless Networks

In this paper, the problem of audio based semantic communication is investigated over wireless networks. In the considered model, wireless edge devices must transmit large-sized audio data to a server using semantic communication techniques. The techniques enable the transmission of audio semantic i...

Full description

Saved in:
Bibliographic Details
Published in2021 IEEE Global Communications Conference (GLOBECOM) pp. 1 - 6
Main Authors Tong, Haonan, Yang, Zhaohui, Wang, Sihua, Hu, Ye, Saad, Walid, Yin, Changchuan
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.12.2021
Subjects
Audio based semantic communication
Collaborative work
federated learning
Mean square error methods
Performance evaluation
Phase change materials
Semantics
Simulation
Wireless networks
Online AccessGet full text

Cover

More Information
Summary:In this paper, the problem of audio based semantic communication is investigated over wireless networks. In the considered model, wireless edge devices must transmit large-sized audio data to a server using semantic communication techniques. The techniques enable the transmission of audio semantic information which captures the contextual features of audio signals. To extract the semantic information from audio signals, a wave to vector (wav2vec) architecture based autoencoder that consists of convolutional neural networks (CNNs) is proposed. The proposed autoencoder enables high-accuracy audio transmission with small amounts of data. To further improve the accuracy of semantic information extraction, federated learning (FL) is implemented over multiple devices and a server. Simulation results show that the proposed algorithm can converge effectively and can reduce the mean square error (MSE) between the recovered audio signals and the source audio signals by nearly 100 times, compared to a traditional coding scheme.
DOI:10.1109/GLOBECOM46510.2021.9685654