In order to achieve remote voice acquisition, a 9.75-GHz microwave Doppler radar with a high-sensitivity coherent homodyne demodulator is used to detect voice signals. However, the combined noise sources, including background noise and shock noise, are included in the Doppler audio radar-captured voice, which seriously damages the quality of speech. In the current work, a multi-stage enhancement method is used to enhance Doppler audio radar-captured voice. The first stage is dictionary learning, which uses the K-singular value decomposition algorithm to train speech and noise dictionaries. The noise training data come from the Doppler audio radar-captured shock noise signals under the influence of different relative motion modulations. The second stage generates a position mask of the shock noise. The third stage is the removal of background noise and shock noise. The removal of background noise adopts our proposed constrained low-rank sparse matrix decomposition algorithm based on power spectral density. The final speech spectrum with shock noise removed is obtained through sparse restoration using the LARC algorithm. The experimental results show that the 9.75-GHz Doppler audio radar can effectively acquire human voice signals, and the proposed multi-stage enhancement method can effectively remove the background noise and shock noise in the radar speech, thereby improving the quality of the audio radar voice. The obtained results imply the potential to implement practical Doppler audio radar systems for human voice signal acquisition.

In recent years, neural networks based on attention mechanisms have seen increasingly use in speech recognition, separation, and enhancement, as well as other fields. In particular, the convolution-augmented transformer has performed well, as it can combine the advantages of convolution and self-attention. Recently, the gated attention unit (GAU) was proposed. Compared with traditional multi-head self-attention, approaches with GAU are effective and computationally efficient. In this CGA-MGAN: MetricGAN based on Convolution-augmented Gated Attention for Speech Enhancement, we propose a network for speech enhancement called CGA-MGAN, a kind of MetricGAN based on convolution-augmented gated attention. CGA-MGAN captures local and global correlations in speech signals at the same time by fusing convolution and gated attention units. Experiments on Voice Bank + DEMAND show that our proposed CGA-MGAN model achieves excellent performance (3.47 PESQ, 0.96 STOI, and 11.09 dB SSNR) with a relatively small model size (1.14 M).

The neural network has become a new and powerful paradigm in speech enhancement, triggering the surge of research. However, most existing methods directly predict speech and ignore the noise information. In this paper, we propose a two-stage learning and fusion network with noise awareness for time-domain monaural speech enhancement, which can be regarded as a progressive learning process. More specifically, in the first network, speech and noise are estimated simultaneously. The estimated speech and the signal by subtracting estimated noise from the noisy speech are stacked with noisy speech as input to obtain further refined speech in the second network. Both networks are mainly built on encoders and decoders with skip connections. To better control the information flow in the network, we introduce the gated linear unit in the encoder and the decoder, which also can help model complex interactions. Dilated dense blocks are added after each layer of the encoder and decoder to improve the model efficiency and enlarge the receptive field. Our experiments confirm that the proposed two-stage learning network with noise awareness achieves better performance than several advanced systems under various conditions.

In recent years, neural networks based on attention mechanisms have been increasingly widely used in speech recognition, separation, enhancement, and other fields. In particular, the convolution-augmented transformer has achieved good performance as it can combine the advantages of convolution and self-attention. Recently, the gated attention unit (GAU) has been proposed. Compared with the traditional multi-head self-attention, approaches with GAU are effective and computationally efficient. In this article, we propose a network for speech enhancement called CGA-MGAN, a kind of Metric GAN based on convolution-augmented gated attention. CGA-MGAN captures local and global correlations in speech signals at the same time through the fusion of convolution and gated attention units. Experiments on Voice Bank + DEMAND show that the CGA-MGAN we propose achieves an excellent performance (3.47 PESQ, 0.96 STOI, and 11.09dB SSNR) at a relatively small model size (1.14M).

Speech enhancement based on deep neural networks faces difficulties, as modeling more frequency bands can lead to a decrease in the resolution of low-frequency bands and increase the computational complexity. Previously, we proposed a convolution-augmented gated attention unit (CGAU), which captured local and global correlation in speech signals through the fusion of the convolution and gated attention unit. In this paper, we further improved the CGAU, and proposed a two-stage complex and real dual-path sub-band fusion network for full-band speech enhancement called TS-CGANet. Specifically, we proposed a dual-path CGA network to enhance low-band (0–8 kHz) speech signals. In the medium band (8–16 kHz) and high band (16–24 kHz), noise suppression is only performed in the magnitude domain. The Voice Bank+DEMAND dataset was used to conduct experiments on the proposed TS-CGANet, which consistently outperformed state-of-the-art full-band baselines, as evidenced by the results.

Magnetic anomaly detection from aerial platforms enables passive, non-contact detection of underwater ferromagnetic targets. To address the issues of traditional Orthonormal Basis Function (OBF) detection relying on prior knowledge of the closest contact distance and susceptibility to mismatch with single-scale templates, this paper proposes an aerial magnetic anomaly target detection method that integrates multi-scale OBF with a convolution-augmented gated attention unit (CGAU). Six sets of fixed OBF convolution kernels each output three matching responses, which are concatenated with the original sequence to form 19-channel features. Four layers of one-dimensional (1-D) convolution and two stages of CGAU extract local waveform features and capture long-range dependencies. The precision (PC), detection rate (DR), and F1-score of the proposed network on the simulated test set are 94.70%, 95.11%, and 94.90%, respectively (best-performing seed = 42). On the measured dataset, it detected 11 of 12 overpass events, achieving an event detection rate of 91.67%. The proposed network has approximately 0.43 M trainable parameters and achieves a higher mean F1 across three random seeds than the seven selected baselines, demonstrating its potential for aerial magnetic anomaly target detection.