Effective integration of available resources within edge nodes is essential to improve the performance of vehicular edge computing (VEC) to support various randomly offloaded tasks with limited computing capacity and ...
详细信息
Fog computing (FC) is a distributed paradigm that complements cloud computing in terms of service delivery. By extending storage and computation of the network edge, fog systems enable location awareness and mobility ...
详细信息
Speech Enhancement is significantly applied in speech processing, as a foundation for downstream tasks. Nowadays, neural networks are well applied in speech enhancement. However, there remain considerable difficulties...
详细信息
ISBN:
(纸本)9789819755905;9789819755912
Speech Enhancement is significantly applied in speech processing, as a foundation for downstream tasks. Nowadays, neural networks are well applied in speech enhancement. However, there remain considerable difficulties for neural networks to improve speech quality. Firstly, existing methods have the problem of speech over-suppression. Because they have not yet taken into account that neural networks influence not only background noise but also clean speech during enhancement. This issue can negatively impact the following tasks. Secondly, striking a balance between model complexity and performance is crucial, especially when deploying on resource-constrained hardware. Existing models often prioritize performance, overlooking the issue of complexity. To solve the problems above, we propose a novel Generative Adversarial Network based on Two-Stage Mask Transformer and information Interaction (TSMGAN-II), consisting of an attention encoder, a two-stage mask transformer, and a dual-feature decoder with information interaction. It effectively captures and models both amplitude and spectral characteristics within the time-frequency domain. Experiments on the VoiceBank+DEMAND dataset show that our model, with 1.39 million parameters, achieves state-of-the-art performance with PESQ of 3.40 and SSNR of 11.81. Moreover, we also introduce a lightweight model with just 0.59M parameters, achieving 97% of the performance of SOTA models with PESQ of 3.31 and SSNR of 11.53.
Artificial intelligence (AI) is the bedrock of management systems and will continue to support human society both now and in the future. These systems' decision-making processes must be transparent, especially in ...
详细信息
it is a revolutionary Structural Vector automobile regression (SV AR) approach for the cause of monitoring and predicting congested networks. The proposed technique utilizes community congestion-conscious prior data e...
详细信息
Autonomous driving is becoming the most important implementation scenario of deep learning and a key technology affecting the industry. Sensors are the key to perception of the external world in autonomous driving sys...
详细信息
The Wuhan market in China is the source of a terrible and undetectable threat to the entire planet. The time between the outbreak and the epidemic was barely a few months. Visual data analysis makes it easier to get i...
详细信息
This paper introduces suitable methods for detecting SMS spam in various forms of the Vietnamese language. The researchers conducted experiments using five algorithms: SVM, Naive Bayes, Random Forests, CNN, and LSTM, ...
详细信息
For the classification of huge amount of textual data we need to classify them for the purpose of information retrieval, document classification, spam filtering and so on. To apply machine learning algorithms we need ...
详细信息
Artificial Intelligence and Machine Learning (AI/ML) are the emulation of human intelligence by computer systems. The AI/ML models have made inroads into Computer vision and Quantum computing due to their tremendous a...
详细信息
暂无评论