Advanced automatic pronunciation error detection (APED) algorithms are usually based on state-of-the-art automatic speech recognition (ASR) techniques. With the development of deep learning technology, end-to-end ASR technology has gradually matured and achieved positive practical results, which provides us with a new opportunity to update the APED algorithm. We first constructed an end-to-end ASR system based on the hybrid connectionist temporal classification and attention (CTC/attention) architecture. An adaptive parameter was used to enhance the complementarity of the connectionist temporal classification (CTC) model and the attention-based seq2seq model, further improving the performance of the ASR system. After this, the improved ASR system was used in the APED task of Mandarin, and good results were obtained. This new APED method makes force alignment and segmentation unnecessary, and it does not require multiple complex models, such as an acoustic model or a language model. It is convenient and straightforward, and will be a suitable general solution for L1-independent computer-assisted pronunciation training (CAPT). Furthermore, we find that in regards to accuracy metrics, our proposed system based on the improved hybrid CTC/attention architecture is close to the state-of-the-art ASR system based on the deep neural network–deep neural network (DNN–DNN) architecture, and has a stronger effect on the F-measure metrics, which are especially suitable for the requirements of the APED task.

The context-related semantic information of the text in the traditional short text classification algorithms are not fully captured, a text classification model based on bidirectional temporal convolutional network and attention mechanism (BTCA) was proposed. Multi-layer dilated convolution was used to increase the receptive field and better capture long distance dependent information. At the same time, attention mechanism was used to increase the attention to the local key feature in the text, and the bidirectional temporal convolution network was used to extract contextual multi-scale semantic information to enrich semantics, the problem of sparse short text features was solved to a certain extent, and text classification effect was improved. The public corpus of the THUCNews was used to conduct a comparative experiment. It is pointed out that effect of short text classification is improved by using BTCA model, with an accuracy rate of 91.47%, which is better than commonly used models.

Early diagnosis of Alzheimer’s disease (AD) and its precursor, mild cognitive impairment (MCI), is critical for effective prevention and treatment. Computer-aided diagnosis using magnetic resonance imaging (MRI) provides a cost-effective and objective approach. However, existing methods often segment 3D MRI images into 2D slices, leading to spatial information loss and reduced diagnostic accuracy. To overcome this limitation, we propose TA-SSM Net, a deep learning model that leverages tri-directional attention and structured state-space model (SSM) for improved MRI-based diagnosis of AD and MCI. The tri-directional attention mechanism captures spatial and contextual information from forward, backward, and vertical directions in 3D MRI images, enabling effective feature fusion. Additionally, gradient checkpointing is applied within the SSM to enhance processing efficiency, allowing the model to handle whole-brain scans while preserving spatial correlations. To evaluate our method, we construct a dataset from the Alzheimer’s Disease Neuroimaging Initiative (ADNI), consisting of 300 AD patients, 400 MCI patients, and 400 normal controls. TA-SSM Net achieved an accuracy of 90.24% for MCI detection and 95.83% for AD detection. The results demonstrate that our approach not only improves classification accuracy but also enhances processing efficiency and maintains spatial correlations, offering a promising solution for the diagnosis of Alzheimer’s disease.

Image blur, often caused by camera shake and object movement, poses a significant challenge in computer vision. Image deblurring strives to restore clarity to these images. Traditional single-stage methods, while effective in detail enhancement, often neglect global context in favor of local information. Yet, both aspects are crucial, especially in real-life scenarios where images are typically large and subject to various blurs. Addressing this, we introduce CNB Net, an innovative deblurring network adept at integrating global and local insights for enhanced image restoration. The network operates in two stages, utilizing our specially designed Convolution and Normalization-Based Block (CNB Block) and Convolution and Normalization-Based Plus Block (CNBP Block) for multi-scale information extraction. A progressive learning approach is adopted with a Feature Active Selection (FAS) module at the end of each stage that captures spatial detail information under the guidance of real images. The Two-Stage Feature Fusion (TSFF) module reduces information loss caused by downsampling operations while enriching features across stages for increased robustness. We conduct experiments on the GoPro dataset and the HIDE dataset. On the GoPro dataset, our Peak Signal-to-Noise Ratio (PSNR) result is 32.21 and the Structural Similarity (SSIM) result is 0.950; and on the HIDE dataset, our PSNR result is 30.38 and the SSIM result is 0.932. Our results exceed other similar algorithms. By comparing the generated feature maps, we find that our model takes into account both global and local information well.

A multi-task learning network traffic classification model (MTL-NTC) is experimentally studied. The identification task of the MTL-NTC model contains three tasks: identifying the type of traffic, predicting the bandwidth, and the duration of the flow. Identifying the type of traffic is the main task. It includes the classification of five types of traffic. The bandwidth prediction and the duration of the flow are auxiliary tasks, which are used to improve the performance of the main task. In the public network traffic data set QUIC, the MTL-NTC model and the common model are used for comparative experiments. The experimental results show that the recognition accuracy of this model is 97.33%, which is better than the common models.

Deep learning-based image deblurring techniques have made great advancements, improving both processing speed and deblurring efficacy. However, existing methods still face challenges when dealing with complex blur types and the semantic understanding of images. The segment anything model (SAM), a versatile deep learning model that accurately and efficiently segments objects in images, facilitates various tasks in computer vision. This article leverages SAM’s proficiency in capturing object edges and enhancing image content comprehension to improve image deblurring. We introduce the edge-sensitive focusing encoder (EFE) module, which utilizes masks generated by the SAM framework and re-weights the masked portion following SAM segmentation by detecting its features and high-frequency information. The EFE module uses the masks to locate the position of the blur in an image while identifying the intensity of the blur, allowing the model to focus more accurately on specific features. Masks with greater high-frequency information are assigned higher weights, prompting the network to prioritize them during processing. Based on the EFE module, we develop a deblurring network called the edge-sensitive focusing encoder-based convolution–normalization and attention network (EFE-CNA Net), which utilizes the EFE module to enhance the deblurring process, employs an image-mask decoder to merge features from both the image and the mask from the EFE module, and incorporates the CNA Net as its base network. This design enables the model to focus on distinct features at various locations, enhancing its learning process through the guidance provided by the EFE module and the blurred images. Testing results on the RealBlur and REDS datasets demonstrate the effectiveness of the EFE-CNA Net, achieving peak signal-to-noise ratio (PSNR) and structural similarity (SSIM) metrics of 28.77, 0.902 (RealBlur-J), 36.40, 0.956 (RealBlur-R), 31.45, and 0.919 (REDS).