While modern deep learning-based models have significantly outperformed traditional methods in the area of speech enhancement, they often necessitate a lot of parameters and extensive computational power, making them impractical to be deployed on edge devices in real-world applications. In this paper, we introduce Grouped Temporal Convolutional Recurrent Network (GTCRN), which incorporates grouped strategies to efficiently simplify a competitive model, DPCRN. Additionally, it leverages subband feature extraction modules and temporal recurrent attention modules to enhance its performance. Remarkably, the resulting model demands ultralow computational resources, featuring only 23.7 K parameters and 39.6 MMACs per second. Experimental results show that our proposed model not only surpasses RNNoise, a typical lightweight model with similar computational burden, but also achieves competitive performance when compared to recent baseline models with significantly higher computational resources requirements.

Although deep learning (DL) based end-to-end models have shown outstanding performance in multi-channel speech extraction, their practical applications on edge devices are restricted due to their high computational complexity. In this paper, we propose a hybrid system that can more effectively integrate the generalized sidelobe canceller (GSC) and a lightweight post-filtering model under the assistance of spatial speaker activity information provided by a directional voice activity detection (DVAD) module. In addition to guiding the update of the adaptive blocking matrix (ABM) and the adaptive interference canceller (AIC) used in GSC to alleviate the distortion of the desired speech, DVAD is also utilized as an auxiliary input to the post-filtering model to enhance its capability of interference suppression. The experimental results demonstrate that, with much lower computational costs, our method can achieve comparable performance with a current state-of-the-art end-to-end model on simulated data and generalize even better on real-world data.

MetricGAN and its variations have been proven to be an effective wide-band speech enhancement model. In this paper, we expand it to full-band enhancement by combining our recently proposed learnable spectral dimension compression mapping strategy. The encoder-decoder structure with a time-frequency convolutional recurrent network is utilized as the generator. The proposed model is submitted to the ICASSP Signal Processing Grand Challenge: DNS-5 Challenge (2023). Without using the enrollment speech, it obtains a final score of 0.548 on Track-1 and 0.559 on Track-2.

This paper summarizes a hybrid multi-channel speech enhancement system for the ICASSP Signal Processing Grand Challenge: Clarity Challenge (Speech Enhancement for Hearing Aids) 2023. The system consists of a rule-based dereverberation module, a multi-channel enhancement module, and a post-processing module. Without using the head rotation information and the enrollment speech, the system can reach an average hearing aid speech perception index (HASPI) score of 0.696 and hearing aid speech quality index (HASQI) score of 0.320 on the official development set. The corresponding scores are 0.729 and 0.316 respectively on the Eval1 set for the challenge ranking.

Compressed sensing (CS) based synthetic aperture radar (SAR) imaging methods are showing superior potential in imaging performance over classical matched filtering based methods. However, the CS-based methods require much more computational cost to solve the iterative optimization composed of large-scale matrix operators. To hold the improvement of imaging performance and reduce the computational cost, in this paper, we propose a novel SAR imaging method by Deep Unfolded Network (DUN) of Iterative Shrinkage Threshold Algorithm (ISTA) with the approximated observation of Range-Doppler Algorithm (RDA) operator. The proposed method takes the radar echoes as the input to learn the imaging procedure. Firstly, the approximated observation is utilized in SAR imaging model to reduce the size of the DUN. Moreover, we use ISTA as an example to introduce how to establish DUN with approximated observation, in which the detailed structure to handle the complex-valued radar echoes is also designed. Finally, the auto-encoder is utilized to calculate the difference of the echoes rather than the imaging results so that we can train the proposed network by unsupervised learning. The experiments of both point targets, surface targets, and real scenes show that the proposed imaging method is superior in terms of imaging performance and computing efficiency.