Design of a Speech Enhancement-Based Deep Semantic Autoencoder for Wireless Communications
Authors: Xiaoling Han, Bin Lin, Nan Wu, Tony Q. S. Quek
Organizations: Department of Information Science and Technology of Dalian Maritime University, Dalian, China, 116026 · Information Systems Technology and Design Pillar, Singapore University of Technology and Design, Singapore, 487372
Speech communication plays a crucial role in the Internet of Things and autonomous networks. However, traditional communication systems suffer from insufficient semantic extraction and limited noise suppression capability. To address these issues, we propose a Speech enhancement-based Deep Semantic Autoencoder (S-DSA) communication system for wireless communications. It consists of a dual-branch semantic encoder, a channel encoder, a channel decoder, and two semantic decoders, enabling joint source-channel coding for robust transmission. A key novelty lies in an adaptive semantic noise decoupling mechanism that disentangles target speech semantics from noise representations through an interaction module, thus maintaining semantic consistency under low SNR conditions. Furthermore, a joint training strategy optimizes speech reconstruction and noise estimation, while a lightweight testing strategy reduces complexity by disabling the noise-decoding path. The proposed S-DSA enables more effective separation between speech and noise, and yields higher robustness as well as better reconstruction quality. These findings highlight the potential of the proposed S-DSA as a practical solution for next-generation intelligent wireless communications.
Figures & tables
Fig. 1: Architecture of the proposed S-DSA.
End
Layer Category
Layer Name
Output Dimension
Transmitter
Dual-branch semantic encoder Only the semantic encoder S is shown; the semantic encoder N is identical.
Conv block (Semantic encoder S)
[ B , Ds , T , F ]
Conv block (Semantic encoder S)
[ B , 2Ds , T , F/2 ]
Conv block (Semantic encoder S)
[ B , 4Ds , T , F/4 ]
Conv stack (Semantic encoder S)
[ B , 4Ds , T , F/4 ]
Conv stack (Semantic encoder S)
[ B , T , F/4 , 4Ds ]
SSA (Semantic encoder S)
[ B , T , F/4 , 4Ds ]
TABLE I: Architectural parameters of the proposed S-DSA.
Parameter
Setting
Implementation framework
PyTorch
Optimizer
Adam
Max training epochs
120
Learning rate
Initial 10−3 , minimum 10−4
Batch size
32
Validation split
20%
TABLE II: Simulation parameter settings of the proposed S-DSA.
Fig. 2: Waveform and spectrogram comparisons for speech enhancement.
Fig. 3: Performance metrics versus channel SNR across different channels.
Fig. 4: Performance metrics versus environmental SNR on the AWGN channel.
Fig. 5: Training and validation loss on the AWGN channel.
Fig. 6: Performance metrics versus channel SNR on the AWGN channel under varying learning rates.
Figure 9Figure 10Figure 11
Tony Q. S. Quek (S’98-M’08-SM’12-F’18) received the B.E. and M.E. degrees in electrical and electronics engineering from the Tokyo Institute of Technology in 1998 and 2000, respectively, and the Ph.D. degree in electrical engineering and computer science from the Massachusetts Institute of Technology in 2008. Currently, he is the Associate Provost (AI & Digital Innovation) and Cheng Tsang Man Chair Professor with Singapore University of Technology and Design (SUTD). He also serves as the Director of the Future Communications R&D Programme, and the ST Engineering Distinguished Professor. He is a co-founder of Silence Laboratories and NeuroRAN. His current research topics include wireless communications and networking, network intelligence, non-terrestrial networks, open radio access network, AI-RAN, and 6G. Dr. Quek was honored with the 2008 Philip Yeo Prize for Outstanding Achievement in Research, the 2012 IEEE William R. Bennett Prize, the 2015 SUTD Outstanding Education Awards – Excellence in Research, the 2016 IEEE Signal Processing Society Young Author Best Paper Award, the 2017 CTTC Early Achievement Award, the 2017 IEEE ComSoc AP Outstanding Paper Award, the 2020 IEEE Communications Society Young Author Best Paper Award, the 2020 IEEE Stephen O. Rice Prize, the 2020 Nokia Visiting Professor, the 2022 IEEE Signal Processing Society Best Paper Award, the 2024 IIT Bombay International Award For Excellence in Research in Engineering and Technology, the IEEE Communications Society WTC Recognition Award 2024, and the Public Administration Medal (Bronze). He is an IEEE Fellow, a WWRF Fellow, an AIIA Fellow, and a Fellow of the Academy of Engineering Singapore.
This paper presents a semantic-enhanced receiver framework for transmitting natural language sentences over noisy wireless channels using multiple short block codes. After ASCII encoding, the sentence is divided into segments, each independently encoded with a short block code and transmitted over an AWGN channel. At the receiver, segments are decoded in parallel, followed by a semantic error correction (SEC) model, which reconstructs corrupted segments using language model context. We further propose the semantic list decoding (SLD), which generates multiple candidate reconstructions and selects the best one via weighted Hamming distance, and a semantic confidence-guided HARQ (SHARQ) mechanism that replaces CRC-based error detection with a confidence score, enabling selective segment retransmission without CRC overhead. All modules are designed and trained using bidirectional and auto-regressive transformers (BART). Simulation results demonstrate that the proposed scheme significantly outperforms conventional capacity-approaching short codes and long codes at the same rate. Specifically, SEC provides approximately 0.4 dB BLER gain over plain short-code transmission, while SLD extends this to 0.8 dB. Compared to transmitting the entire sentence as a single long 5G LDPC codeword, our approach significantly improves semantic fidelity and reduces decoding latency by up to 90%. SHARQ further provides an additional 1.5 dB gain over conventional HARQ.
Jiafu Hao, Chentao Yue, Wanchun Liu +2
School of Electrical and Computer Engineering, The University of Sydney, Sydney, NSW 2006, Australia
Internet of Things (IoT) networks face significant challenges such as limited communication bandwidth, constrained computational and energy resources, and highly dynamic wireless channel conditions. Utilization of deep neural networks (DNNs) combined with semantic communication has emerged as a promising paradigm to address these limitations. Deep joint source-channel coding (DJSCC) has recently been proposed to enable semantic communication of images. Building upon the original DJSCC formulation, low-complexity attention-style architectures has been added to the DNNs for further performance enhancement. As a main hurdle, training these DNNs separately for various signal-to-noise ratios (SNRs) will amount to excessive storage or communication overhead, which can not be maintained by small IoT devices. SNR Adaptive DJSCC (ADJSCC), has been proposed to train the DNNs once but feed the current SNR as part of the data to the channel-wise attention mechanism. We improve upon ADJSCC by a simultaneous utilization of doubly adaptive channel-wise and spatial attention modules at both transmitter and receiver. These modules dynamically adjust to varying channel conditions and spatial feature importance, enabling robust and efficient feature extraction and semantic information recovery. Simulation results corroborate that our proposed doubly adaptive DJSCC (DA-DJSCC) significantly improves upon ADJSCC in several performance criteria, while incurring a mild increase in complexity. These facts render DA-DJSCC a desirable choice for semantic communication in performance demanding but low-complexity IoT networks.
Deep learning based semantic communication has achieved significant progress in wireless image transmission, but most existing schemes rely on fixed models and thus lack robustness to diverse image contents and dynamic channel conditions. To improve adaptability, recent studies have developed adaptive semantic communication strategies that adjust transmission or model behavior according to either source content or channel state. More recently, MoE-based semantic communication has emerged as a sparse and efficient adaptive architecture, although existing designs still mainly rely on single-driven routing. To address this limitation, we propose a novel multi-stage end-to-end image semantic communication system for multi-input multi-output (MIMO) channels, built upon an adaptive MoE Swin Transformer block. Specifically, we introduce a dynamic expert gating mechanism that jointly evaluates both real-time CSI and the semantic content of input image patches to compute adaptive routing probabilities. By selectively activating only a specialized subset of experts based on this joint condition, our approach breaks the rigid coupling of traditional adaptive methods and overcomes the bottlenecks of single-driven routing. Simulation results indicate a significant improvement in reconstruction quality over existing methods while maintaining the transmission efficiency.
Haowen Wan, Qianqian Yang
Polytechnic Institute, Zhejiang University, Hangzhou, China · College of Information Science and Electronic Engineering, Zhejiang University, Hangzhou, China