检索结果-内蒙古大学图书馆

International Conference on Acoustics, Speech, and signal Processing (ICASSP)

作者： Ke Chen Zhihua Huang Liang He Yonghong Yan School of Computer Science and Technology Xinjiang University Urumqi China Xinjiang Key Laboratory of Signal Detection and Processing Urumqi China Department of Electronic Engineering Tsinghua University Beijing China University of Chinese Academy of Sciences Beijing China Key Laboratory of Speech Acoustics and Content Understanding Institute of Acoustics CAS Beijing China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Code-Switching (CS) Text-To-Speech (TTS) models have gained attention due to the increasing prevalence of multilingual communication. However, existing models struggle to meet the growing demand for personalized CS TTS, particularly in Zero-Shot (ZS) scenarios where the model should generate speech for an unseen speaker from reference speech. In this paper, we propose ZCS-CDiff, a zero-shot code-switching TTS system with the Conformer-based diffusion model. We disentangle speech features and use the diffusion model to precisely model these disentangled attributes, resulting in high-quality ZS CS speech. Additionally, we introduce a Conformer-based WaveNet as the denoising network within the diffusion model to further improve the accuracy of modeling different attributes. We also designed a speaker-assist module to help the model better handle speaker information extracted from the reference speech, resulting in CS speech with higher speaker similarity. Experimental results and ablation studies demonstrate that the ZS CS speech generated by ZCS-CDiff has good speech naturalness, intelligibility, and speaker similarity while confirming the effectiveness of our design choices. Audio samples are available 1 .

关键词： Training Adaptation models Accuracy Speech coding Noise reduction Speech enhancement signal processing Diffusion models Probabilistic logic Text to speech

来源：评论

学校读者我要写书评

暂无评论

MRLCD-A: Lag-aware alignment for multivariate time series forecasting in multiple scenarios

引用

Information Processing and Management 2025年第5期62卷

作者： Sun, Dezhi Qin, Jiwei Zhang, Zihao Qin, Xizhong Zhang, Huiguo School of Computer Science and Technology Xinjiang University Xinjiang Uygur Autonomous Region Urumqi 830046 China Xinjiang Key Laboratory of Signal Detection and Processing Xinjiang University Urumqi830046 China School of Mathematics and System Science Xinjiang University Xinjiang Uygur Autonomous Region Urumqi 830046 China

In multivariate time series forecasting tasks, the varying degrees of lag relationships among multivariate data significantly increase the complexity of accurate predictions. A model must effectively capture long-term dependencies and address intricate lag correlations to achieve reliable long-term forecasting. This paper proposes a novel Multivariate Rolling Lag Correlation detection-Alignment (MRLCD-A) method to tackle these challenges. The method identifies rolling correlations, calculates lag distances in multivariate sequence inputs, and aligns the lagged variables accordingly. Multivariate Time Series (MTS) forecasting uses a Channel Dependency (CD) approach. Experiments on time series datasets across various scenarios, including electricity, weather, exchange rates, and atmospheric carbon concentrations, demonstrate that the proposed method outperforms state-of-the-art models in forecasting general multivariate time series and predicting long-term time series data in real-world environments. © 2025 Elsevier Ltd

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

MN-Net: Speech Enhancement Network via Modeling the Noise

引用

IEEE Transactions on Audio, Speech and Language Processing 2025年 33卷 1208-1219页

作者： Ying Hu Qin Yang Wenbing Wei Li Lin Liang He Zhijian Ou Wenzhong Yang School of Computer Science and Technology Key Laboratory of Signal Detection and Processing Xinjiang University Urumqi China Xinjiang Institute of Electronics Research Shares Company Ltd. Urumqi China Department of Electrical Engineering Tsinghua University Beijing China Beijing National Research Center for information Science and Technology Department of Electronics Tsinghua University Beijing China

Currently, deep learning-based speech enhancement methods generally focus on target speech extraction while neglecting modeling the other sound sources in the mixture. These methods still can't distinguish the target speech from the interference well. In this paper, we present a monaural speech enhancement network via Modeling the Noise (MN-Net), which includes a shared Encoder and three separate Decoders for parallel modeling the magnitude and phase spectrogram of target speech, and the complex spectrogram of noise. Specifically, we propose a Multi-Branch Feature Extractor (MBFE) module to capture the richer contextual information in mixture, and a Spatial Reconstruction Unit (SRU) to remove the redundancy from extracted features. We compared our proposed MN-Net with 18 classical speech enhancement methods on the VoiceBank+DEMAND dataset, and with 9 ones on DNS-Challenge dataset for denoising task, and with 7 ones on the WHAMR! dataset for simultaneous denoising & de-reverberation task. Our proposed MBFE module was applied to two classical speech enhancement methods, DB-AIAT and CMGAN, replacing their DenseBlocks module. The results demonstrate that applying the MBFE module can boost their performances while keeping smaller model size. A series of visualization analysis intuitively verify that modeling the noise can enable the network to distinguish the target speech from noise and other interference more accurately.

关键词： Noise Feature extraction Spectrogram Speech enhancement Decoding Kernel Transformers Noise reduction Data mining Training

来源：评论

学校读者我要写书评

暂无评论

TOEformer: Temporal order enhanced Transformer for time series forecasting

引用

Engineering Applications of Artificial Intelligence 2025年 155卷

作者： Li, Qiang Qin, Jiwei Wang, Dacheng Qin, Xizhong Cui, Daishun Xie, Jiachen Sun, Dezhi School of Computer Science and Technology Xinjiang University Urumqi China Xinjiang Key Laboratory of Signal Detection and Processing Xinjiang University Urumqi China Aerospace Information Research Institute Chinese Academy of Sciences Beijing China

Transformers have shown excellent performance in time series forecasting, but a recent study points out that existing Transformer-based time series forecasting models cannot preserve temporal order well. To address this problem, we preserve temporal order information by integrating linear and Transformer structures and predicting trend and seasonal data, respectively, which compensates for the inadequacy of the Transformer in this regard by leveraging linear models to preserve the temporal order information of the original sequence. Besides, directly combining the seasonal and trend part prediction results increases the overall prediction error. For this problem, this paper proposes a new loss function, the relative value exponential loss, which reduces the impact of the problem on predictions by constraining the network to generate more seasonal-trend pairs that can offset each other's errors. Experimental results on six benchmark public datasets show that our temporal order enhanced Transformer (TOEformer) can preserve temporal order information and produce better forecasting performance in more than 85% of multivariate prediction setups compared to state-of-the-art time series forecasting models. The proposed relative value exponential loss effectively mitigates the overall error growth problem. © 2025

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

A Singing Melody Extraction Network Via Self-Distillation and Multi-Level Supervision

A Singing Melody Extraction Network Via Self-Distillation an...

引用

International Conference on Acoustics, Speech, and signal Processing (ICASSP)

作者： Ying Hu Jiabo Jing Fan Li Lijun He Li Lin Wenzhong Yang Key Laboratory of Signal Detection and Processing Urumqi China School of Computer Science and Technology Xinjiang University Urumqi China School of Information and Communications Engineering Xi’an Jiaotong University Xi’an China Xinjiang Institute of Electronics Research Shares Co. Ltd Urumgi China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Extracting singing melody from polyphonic music is an important topic in the field of music information retrieval. In this paper, we propose a singing melody extraction network consisting of five stacked multi-scale feature time-frequency aggregation (MF-TFA) modules. In the same network, deeper layers generally contain more contextual information than shallower layers. To help the shallower layers enhance the ability of task-relevant feature extraction, we propose a self-distillation and multi-level supervision (SD-MS) method, which leverages the feature distillation from the deepest layer to the shallower one and multi-level supervision to guide network training. Visualization analysis shows that by introducing SD-MS, the same-level layer in the network can obtain a clearer representation of fundamental frequency components, while the shallower layers can even learn more task-relevant semantic information. Ablation study results indicate that SD-MS applies to existing melody extraction models and can consistently improve performance. Experimental results show that our proposed method, MF-TFA with SD-MS, outperforms six compared state-of-the-art methods, achieving overall accuracy (OA) scores of 87.1%, 89.9%, and 76.6% on the ADC 2004, MIREX 05, and MEDLEY DB datasets, respectively. The main code will be available at https://***/SmoothJing/MF-TFA_SD-MS.

关键词： Training Visualization Time-frequency analysis Semantics signal processing Feature extraction Multiple signal classification Data mining Speech processing Music information retrieval

来源：评论

学校读者我要写书评

暂无评论

Stable Extended U-Net for Noise-Robust Speaker Verification

Stable Extended U-Net for Noise-Robust Speaker Verification

引用

International Conference on Acoustics, Speech, and signal Processing (ICASSP)

作者： Zonghui Wang Zhihua Fang Liang He School of Computer Science and Technology Xinjiang University Urumqi China Xinjiang Key Laboratory of Signal Detection and Processing Urumqi China School of Intelligence Science and Technology Xinjiang University Urumqi China Department of Electronic Engineering Tsinghua University Beijing China Beijing National Research Center for Information Science and Technology Tsinghua University Beijing China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

With advancements in deep learning, speaker verification systems have significantly improved their performance in noisy environments. Researchers typically demonstrate the effectiveness of their improved models by comparing performance on specific datasets, such as the VoxCeleb benchmark. However, in diverse real-world noise conditions, the out-of-domain generalization ability is also a crucial factor in evaluating a model’s performance improvement. Research on stable learning indicates that eliminating the spurious correlation between training and testing data can enhance the generalization of the model. Building on this idea, we propose an improved speaker verification system with high generalization based on the extended U-Net (ExU-Net). It uses the sample reweighting method from stable learning to eliminate sample correlations and retains more effective speaker information through subpixel convolutions and coordinate attention mechanisms. We validate the effectiveness of this approach through extensive evaluations on VoxCeleb1, VOiCES, and other out-of-domain noise test sets, highlighting its generalization capability and model robustness.

关键词： Training Correlation Attention mechanisms Convolution Data models Stability analysis Robustness Noise measurement Speech processing signal to noise ratio

来源：评论

学校读者我要写书评

暂无评论

Research on cancer rapid diagnosis technology based on adaptive channel attention mechanism and two-dimensional spectral fusion

引用

Microchemical Journal 2025年 213卷

作者： Liu, Wei Chang, Chenjie Li, Jiahe Lv, Xiaoyi Liu, Yang Chen, Chen Chen, Cheng College of Software Xinjiang University Urumqi 830046 China School of Computer Science and Technology Xinjiang University Urumqi 830046 China Key Laboratory of Signal Detection and Processing Xinjiang University Urumqi 830046 China Xinjiang Cloud Computing Application Laboratory Xinjiang Cloud Computing Engineering Technology Research Center Karamay 834000 China

With the widespread use of Artificial Intelligence (AI) in biomedicine, integrating infrared and Raman spectra with biological information offers new prospects for enhanced disease diagnosis. For these two types of data, more information can be extracted using multimodal data fusion compared to single modality, but existing studies tend to only fuse one-dimensional spectral data, ignoring the richer information of the spectra when they are converted to two-dimensional images. In this study, we propose a rapid diagnostic method for cancer, which first converts one-dimensional (1D) Raman and Infrared spectra into two-dimensional (2D) spectral images using Short-Time Fourier Transform (STFT), respectively. Then the features are fused by our designed Adaptive Graph Channel Attention Spacial Cross-Attention Fusion Module (ASCFM). This fusion module innovatively uses the adaptive graph channel attention mechanism to strengthen the model's ability to capture the key information of different spectral features, and utilizes the spatial attention fusion module and the cross-attention module for spatial information extraction and fusion. The output fused features effectively avoid feature information loss through this multimodal feature extraction and fusion, which greatly improves the performance of the model. In this study, Gradient-weighted Class Activation Mapping is also used to make the decision-making process of the neural network more interpretable. The results show that the accuracy of 2D Raman spectral images fused with 2D infrared spectral images by ASCFM followed by ResidualNetwork (ResNet) for diagnosis of lung cancer and glioma datasets is 98.88 % and 96.38 %, which suggests that it is a highly promising adjunctive technique for cancer diagnosis. © 2025

关键词： Auxiliary diagnostic technology Deep learning Infrared spectroscopy Lung cancer Multimodal fusion Raman spectroscopy

来源：评论

学校读者我要写书评

暂无评论

Diagnosis of array antennas based on near-field data using Faster R-CNN

引用

Materials and Design 2025年 254卷

作者： Yang, Boguang Wei, Yulun Shi, Jixiang Hong, Tao Li, Liangyu Xu, Kai-Da National Key Laboratory of Radar Detection and Sensing Xidian University Xi'an710071 China Hangzhou Institute of Technology Xidian University Hangzhou311200 China School of Microelectronics Faculty of Electronic and Information Engineering Xi'an Jiaotong University Xi'an710049 China Department of Signal Theory and Communications Polytechnic School University of Alcalá Alcalá de Henares28871 Spain

In this paper, a source reconstruction method for detecting failures in array antenna elements using near-field data based on Faster region-convolutional neural network (Faster R-CNN) is introduced. The proposed method leverages the convergence and optimization of deep learning to eliminate the ill-posedness of the inverse equation of the electric field integral equation (EFIE). The near-field pattern images of an array are collected, divided into training and test datasets, and then fed into the Faster R-CNN to learn the inverse operator of the EFIE formula. Four different excitation states are trained using the same neural network to compare the effects of excitation faults on feature extraction and classification accuracy. A prototype of the array antenna is fabricated and experimentally verified, with good detection results. Simulation and measurement results show that the four types of array excitation state can be classified and located, and with a mean average precision (mAP) of 93.57%, a training time of 24 minutes, and an efficiency of 0.14 seconds per image. The proposed method realizes the classification and location diagnosis of the excitation faults of array antenna with low computing resource requirements, providing a high efficiency and accuracy evaluation method for the microwave array. © 2025 The Authors

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Feature-Consistent Cyclic Domain Adversarial Network for Unsupervised Transfer Diagnosis of the Autoimmune Disease Igg4-Rd by Raman Spectroscopy

SSRN

引用

SSRN 2025年

作者： Zhang, Ziyang Chen, Cheng Qin, Wentao Wu, Xue Liu, Yang Wu, Lijun Lv, Xiaoyi Chang, Chenjie Chen, Xinya Chen, Zhiyuan Xu, Qianxi Li, Zhuoyu Ma, Mengsi Chen, Chen Xinjiang University College of Software Urumqi830046 China Xinjiang Cloud Computing Application Laboratory Xinjiang University Karamay834099 China School of Computer Science and Technology Xinjiang University Urumqi830046 China Key Laboratory of signal detection and processing Xinjiang University Urumqi830046 China Department of Rheumatology＆Immunology People's Hospital of Xinjiang Uygur Autonomous Region Xinjiang Urumqi China Xinjiang Clinical Research Center for Rheumatoid arthritis Xinjiang Urumqi China

Abstracts IgG4-related disease (IgG4-RD) is a multi-organ immune disorder characterized by systemic involvement, diverse pathogenesis, and rarity, which complicates its diagnosis. Traditional medical diagnostic models, predominantly based on supervised learning in machine learning or deep learning frameworks, struggle to address the scarcity of labeled data due to the diagnostic challenges of IgG4-RD. To overcome this limitation, this study proposes a novel framework based on Feature-Consistent Cyclic Domain Adversarial Networks (FCCDAN). The framework first trains a classification model on the source domain for diagnosing source domain diseases under supervised conditions. Subsequently, leveraging the source domain model, the encoder is guided to extract semantic features, while feature consistency loss is integrated into both the generator and discriminator to align the semantic features of the target domain with those of the source domain. Finally, the aligned features are processed through a pre-trained linear mapping to achieve unsupervised classification and diagnosis in the target domain. The proposed model utilizes common autoimmune diseases as a knowledge source, enabling unsupervised transfer diagnosis for IgG4-RD, and demonstrates generalizability through experiments on other disease datasets, achieving an average diagnostic accuracy of 90.62%. This represents improvements of 4.14%, 9.82%, and 10.51% over baseline models, respectively. FCCDAN expands the applicability of Raman spectroscopy in disease diagnosis, maintaining high performance even in label-scarce environments. The overall research workflow also establishes a robust paradigm for the application and advancement of domain adaptation techniques in this context. © 2025, The Authors. All rights reserved.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

A Lightweight Speech Emotion Recognition Model with Bias-Focal Loss 6

A Lightweight Speech Emotion Recognition Model with Bias-Foc...

引用

6th International Conference on Information Communication and signal Processing, ICICSP 2023

作者： Hou, Shijing Yang, Huamin Hu, Ying Xinjiang University Key Laboratory of Signal Detection and Processing Urumqi China

ISBN: (纸本)9798350339994

A major challenge in speech emotion recognition (SER) is how to build a lightweight model with limited training data for applying on the devices with limited in resources. In this paper, we propose a lightweight speech emotion recognition (SER) model with Bias-Focal loss function, where a Dynamic Separable Convolution (DySC) block is designed to extract more fine-grained emotional features and makes the model smaller. We propose a Bias-Focal loss to address the issue of the inconsistent of training samples, while focusing on data points with high feature diversity during the training phase. Experimental results show that our proposed lightweight model is the smallest comparing with other methods, while the number of parameters in our proposed model is 0.85M. Meanwhile, our proposed model achieves the best performance that the score of Unweighted Accuracy (UA) is 75.01 %, and that of Weighted Accuracy (WA) and F1-score are 74.05% and 74.29 % on the IEMOCAP (scripted+improvised)dataset. © 2023 IEEE.

关键词： Bias-Focal loss dynamic separable convolution lightweight Speech emotion recognition

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：