Natural Language Speech Research Articles

Hybrid multi-objective evolutionary model compression with convolutional neural networks

Deep learning has been utilized in the fields of image processing, natural language processing and speech recognition. For improving the structure of deep learning, how to compress Convolutional Neural Networks has become a major focus topic. The study proposes a sparsification-based Convolutional Neural Networks hybrid multi-objective evolutionary model compression method based on hybrid multi-objective evolution, in which the process improves the sparsity of the model by sparse pruning and introduces a time-domain residual module to provide more robust generalization performance, and finally the function of the research method is tested. The test outcomes indicate that the Convolutional Neural Networks model generated by the research method drops to the lowest value in the first 10 iterations; in the sparsity test, the total weight density of the model generated by the research method is about 36%; when the latency test is conducted, the latency of the research method is between 3 ms and 14 ms when the weight sparsity is 60%. The above results show that the research method has better model compression effect and can generate Convolutional Neural Networks models with both model accuracy and model size, which can provide new technical support for model compression of convolutional neural networks.

An overview memristor based hardware accelerators for deep neural network

AbstractThe prevalence of artificial intelligence applications using artificial neural network architectures for functions such as natural language processing, text prediction, object detection, speech, and image recognition has significantly increased in today's world. The computational functions performed by artificial neural networks in classical applications require intensive and large‐scale data movement between memory and processing units. Various software and hardware efforts are being made to perform these operations more efficiently. Despite these efforts, latency in data traffic and the substantial amount of energy consumed in data processing emerge as bottleneck disadvantages of the Von Neumann architecture. To overcome this bottleneck problem, it is necessary to develop hardware units specific to artificial intelligence applications. For this purpose, neuro‐inspired computing chips are believed to provide an effective approach by designing and integrating a set of features inspired by neurobiological systems at the hardware level to address the problems arising in artificial intelligence applications. The most notable among these approaches is memristor‐based neuromorphic computing systems. Memristors are seen as promising devices for hardware‐level improvement in terms of speed and energy because they possess non‐volatile memory and exhibit analog behavior. They enable effective storage and processing of synaptic weights, offering solutions for hardware‐level development. Taking into account these advantages of memristors, this study examines the research conducted on artificial neural networks and hardware that can directly perform deep learning functions and mimic the biological brain, which is different from classical systems in today's context.

Development of the Multimodal Handling Interface Based on Google API

Today, Artificial Intelligence is a daily routine, becoming deeply entrenched in our lives. One of the most popular and rapidly advancing technologies is speech recognition, which forms an integral part of the broader concept of multimodal data handling. Multimodal data encompasses voice, audio, and text data, constituting a multifaceted approach to understanding and processing information. This paper presents the development of a multimodal handling interface leveraging Google API technologies. The interface aims to facilitate seamless integration and management of diverse data modalities, including text, audio, and video, within a unified platform. Through the utilization of Google API functionalities, such as natural language processing, speech recognition, and video analysis, the interface offers enhanced capabilities for processing, analysing, and interpreting multimodal data. The paper discusses the design and implementation of the interface, highlighting its features and functionalities. Furthermore, it explores potential applications and future directions for utilizing the interface in various domains, including healthcare, education, and multimedia content creation. Overall, the development of the multimodal handling interface based on Google API represents a significant step towards advancing multimodal data processing and enhancing user experience in interacting with diverse data sources.

A Study of Artificial Intelligence in the Production of Film

As a key branch of the intelligence field, Artificial Intelligence (AI) is a technical science with the goal of understanding the nature of intelligence and inventing new types of intelligent machines that can act similarly to human intelligence. The study of AI technology includes robotics, natural language processing, speech and image recognition, etc. In recent years, AI has been rapidly developed and widely deployed in a variety of applications in different industries such as autonomous driving, healthcare, financial services, and film production. The application of AI in films is studied in this article, and according to analysis, it can be concluded that in the process of film production, AI technology has its advantages in increasing efficiency and cost-effectiveness in film production. It can also enhance the overall quality of films, impressing the audience by generating special effects and providing filmmakers with new tools and techniques to explore and experiment with. Meanwhile, AI technology also has its disadvantages such as a lack of human touch and depth in the the works produced by AI.

Machine learning in healthcare strategic management: a systematic literature review

PurposeThis study specifically seeks to investigate the strategic implementation of machine learning (ML) algorithms and techniques in healthcare institutions to enhance innovation management in healthcare settings.Design/methodology/approach The papers from 2011 to 2021 were considered following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. First, relevant keywords were identified, and screening was performed. Bibliometric analysis was performed. One hundred twenty-three relevant documents that passed the eligibility criteria were finalized.Findings Overall, the annual scientific production section results reveal that ML in the healthcare sector is growing significantly. Performing bibliometric analysis has helped find unexplored areas; understand the trend of scientific publication; and categorize topics based on emerging, trending and essential. The paper discovers the influential authors, sources, countries and ML and healthcare management keywords.Research limitations/implications The study helps understand various applications of ML in healthcare institutions, such as the use of Internet of Things in healthcare, the prediction of disease, finding the seriousness of a case, natural language processing, speech and language-based classification, etc. This analysis would help future researchers and developers target the healthcare sector areas that are likely to grow in the coming future.Practical implications The study highlights the potential for ML to enhance medical support within healthcare institutions. It suggests that regression algorithms are particularly promising for this purpose. Hospital management can leverage time series ML algorithms to estimate the number of incoming patients, thus increasing hospital availability and optimizing resource allocation. ML has been instrumental in the development of these systems. By embracing telemedicine and remote monitoring, healthcare management can facilitate the creation of online patient surveillance and monitoring systems, allowing for early medical intervention and ultimately improving the efficiency and effectiveness of medical services.Originality/value By offering a comprehensive panorama of ML's integration within healthcare institutions, this study underscores the pivotal role of innovation management in healthcare. The findings contribute to a holistic understanding of ML's applications in healthcare and emphasize their potential to transform and optimize healthcare delivery.

Microsoft's Layout in the Field of Speech Recognition: Taking the Acquisition of Nuance as an Example

This case analysis examines the acquisition of Nuance Communications by Microsoft, with a specific focus on its implications and impact on both companies and the industry. The study reveals that Microsoft's strategic acquisition provides crucial support for its advancements in artificial intelligence and healthcare technology, thereby bolstering its capabilities in medical natural language processing, speech recognition, and electronic health records. Consequently, this acquisition strengthens Microsoft's position in the healthcare sector. For Nuance, the acquisition offers access to greater resources and market channels, facilitating the promotion of its healthcare technologies and solutions. The research emphasizes the significance of this acquisition in comprehending trends within AI and healthcare technology, while exemplifying the positive impact of mergers on driving industry innovation and collaboration. Investors are advised to closely monitor developments in AI and healthcare technology and seize investment opportunities, while leaders should actively foster collaboration and innovation to enhance healthcare services through AI applications. Overall, the acquisition illustrates the potential of this sector and emphasizes the need for strategic positioning in the evolving technological landscape.

Source File Tracking Localization: A Fault Localization Method for Deep Learning Frameworks

Deep learning has been widely used in computer vision, natural language processing, speech recognition, and other fields. If there are errors in deep learning frameworks, such as missing module errors and GPU/CPU result discrepancy errors, it will cause many application problems. We propose a source-based fault location method, SFTL (Source File Tracking Localization), to improve the fault location efficiency of these two types of errors in deep learning frameworks. We screened 3410 crash reports on GitHub and conducted fault location experiments based on those reports. The experimental results show that the SFTL method has a high accuracy, which can help deep learning framework developers quickly locate faults and improve the stability and reliability of models.

Development of Mathematical Models for Industrial Processes Using Dynamic Neural Networks

Dynamic neural networks (DNNs) are a type of artificial neural network (ANN) designed to work with sequential data where context in time is important. Unlike traditional static neural networks that process data in a fixed order, dynamic neural networks use information about past inputs, which is important if the dynamic of a certain process is emphasized. They are commonly used in natural language processing, speech recognition, and time series prediction. In industrial processes, their use is interesting for the prediction of difficult-to-measure process variables. In an industrial isomerization process, it is crucial to measure the quality attributes that affect the octane number of gasoline. Process analyzers commonly used for this purpose are expensive and subject to failure. Therefore, to achieve continuous production in the event of a malfunction, mathematical models for estimating product quality attributes are imposed as a solution. In this paper, mathematical models were developed using dynamic recurrent neural networks (RNNs), i.e., their subtype of a long short-term memory (LSTM) architecture. The results of the developed models were compared with the results of several types of other data-driven models developed for an isomerization process, such as multilayer perceptron (MLP) artificial neural networks, support vector machines (SVM), and dynamic polynomial models. The obtained results are satisfactory, suggesting a good possibility of application.

Artificial intelligence-based language learning: illuminating the impact on speaking skills and self-regulation in Chinese EFL context.

This study investigated the effectiveness of artificial intelligence-based instruction in improving second language (L2) speaking skills and speaking self-regulation in a natural setting. The research was conducted with 93 Chinese English as a foreign language (EFL) students, randomly assigned to either an experimental group receiving AI-based instruction or a control group receiving traditional instruction. The AI-based instruction leveraged the Duolingo application, incorporating natural language processing technology, interactive exercises, personalized feedback, and speech recognition technology. Pre- and post-tests were conducted to assess L2 speaking skills and self-regulation abilities. The results of the study demonstrated that the experimental group, which received AI-based instruction, exhibited significantly greater improvement in L2 speaking skills compared to the control group. Moreover, participants in the experimental group reported higher levels of self-regulation. These findings suggest that AI-based instruction effectively enhances L2 speaking skills and fosters self-regulatory processes among language learners, highlighting the potential of AI technology to optimize language learning experiences and promote learners' autonomy and metacognitive strategies in the speaking domain. However, further research is needed to explore the long-term effects and specific mechanisms underlying these observed improvements.

Unleashing the Power of Deep Attention Networks: A Comprehensive Approach for Enhanced Artificial Intelligence

Deep learning has revolutionized the field of artificial intelligence by achieving state-of-the-art performance on a variety of complex tasks. Attention mechanisms have emerged as a powerful tool to enhance the capabilities of deep neural networks by enabling them to selectively focus on relevant information. In this article, we propose a novel artificial intelligence algorithm called Deep Attention Networks (DANs), which associate multiple attention mechanisms to improve performance on interesting tasks. We evaluate DANs on benchmark datasets in natural language processing, computer vision, and speech recognition and demonstrate superior results compared to existing state-of-the-art approaches. Our approach opens up new possibilities for advancing the field of artificial intelligence and holds promise for various real-world applications. Overall, our results demonstrate the effectiveness and potential of DANs for various AI applications, and highlight the power of combining deep neural networks with attention mechanisms.

Few-Shot Learning Approaches for Fault Diagnosis Using Vibration Data: A Comprehensive Review

Fault detection and diagnosis play a crucial role in ensuring the reliability and safety of modern industrial systems. For safety and cost considerations, critical equipment and systems in industrial operations are typically not allowed to operate in severe fault states. Moreover, obtaining labeled samples for fault diagnosis often requires significant human effort. This results in limited labeled data for many application scenarios. Thus, the focus of attention has shifted towards learning from a small amount of data. Few-shot learning has emerged as a solution to this challenge, aiming to develop models that can effectively solve problems with only a few samples. This approach has gained significant traction in various fields, such as computer vision, natural language processing, audio and speech, reinforcement learning, robotics, and data analysis. Surprisingly, despite its wide applicability, there have been limited investigations or reviews on applying few-shot learning to the field of mechanical fault diagnosis. In this paper, we provide a comprehensive review of the relevant work on few-shot learning in mechanical fault diagnosis from 2018 to September 2023. By examining the existing research, we aimed to shed light on the potential of few-shot learning in this domain and offer valuable insights for future research directions.

Techniques and Applications for Deep Learning: A Review

Deep learning is a branch of machine learning that focuses on the development and refinement of complex neural networks for data analysis, prediction, and decision-making. Deep learning models use numerous layers of artificial neurons to automatically extract important features from raw data, making them superior at many tasks to typical machine learning models. Deep learning models' success in these fields has enhanced state-of-the-art performance and created new research and application prospects. Deep learning has been popular due to its capacity to tackle complicated issues in computer vision, natural language processing, speech recognition, and decision-making. In this study, we discuss deep learning techniques and applications, including recurrent neural networks, long short-term memory, convolutional neural networks, generative adversarial networks, and autoencoders. We also demonstrate deep learning's use in various fields. Deep learning has transformed artificial intelligence by enabling computers to learn from enormous datasets and accomplish complex tasks. As a result, scientists and engineers in fields as diverse as medicine, farming, manufacturing, and transportation have increased their focus on developing deep-learning methods and software. Current research trends and potential future paths in deep learning are also highlighted.

Application of Deep Learning in the Diagnosis of Alzheimer's and Parkinson's disease-A Review.

Most neurodegenerative diseases such as Alzheimer's and Parkinson's are life-threatening, critical, and incurable affecting mainly the elderly population. Early diagnosis is challenging as disease phenotype is very crucial for predicting, preventing the progression, and effective drug discovery. In the last few years, Deep learning (DL) based neural networks are the state-of-the-art models deployed in industries and academics across different areas like natural language processing, image analysis, speech recognition, audio classification, and many more. It has been slowly realized that they have a high potential in medical image analysis and diagnostics and medical management in general. As this field is vast and expanding rapidly, we have put focused on existing DL-based models to detect Alzheimer's and Parkinson's in particular. This study gives a summary of related medical examinations for these diseases. Frameworks and applications of many deep learning models have been discussed. We have given precise notes on pre-processing techniques used by various studies for MRI image analysis. An overview of the application of DL-based models in different stages of medical image analysis has been conferred. It has been realized from the review that more studies are focused on Alzheimer's compared to Parkinson's disease. Additionally, we have tabulated the various public datasets available for these diseases. We have highlighted the potential use of a novel biomarker for the early diagnosis of these disorders. Also, some challenges and issues in implementing deep learning techniques for the detection of these diseases have been addressed. Finally, we concluded with some directions for future research regarding deep learning in the diagnosis of these diseases.

Robust and Adaptive Deep Learning via Bayesian Principles

Deep learning models have achieved tremendous successes in accurate predictions for computer vision, natural language processing and speech recognition applications. However, to succeed in high-risk and safety-critical domains such as healthcare and finance, these deep learning models need to be made reliable and trustworthy. Specifically, they need to be robust and adaptive to real-world environments which can be drastically different from the training settings. In this talk, I will advocate for Bayesian principles to achieve the goal of building robust and adaptive deep learning models. I will introduce a suite of uncertainty quantification methods for Bayesian deep learning, and demonstrate applications en- abled by accurate uncertainty estimates, e.g., robust predic- tion, continual learning and repairing model failures. I will conclude by discussing the research challenges and potential impact for robust and adaptive deep learning models. This paper is part of the AAAI-23 New Faculty Highlights.

Improving a neural network model by explanation-guided training for glioma classification based on MRI data

In recent years, artificial intelligence systems have come to the forefront. These systems, mostly based on deep learning, achieve excellent results in areas such as image processing, natural language processing and speech recognition. Despite the statistically high accuracy of deep learning models, their output is often based on ”black box” decisions. Thus, interpretability methods (Reyes et al. in Radiol Artif Intell 2(3):e190043, 2020) have become a popular way to gain insight into the decision-making process of deep learning models (Miller in Artif Intell 267:1–38, 2019). Explanation of deep learning models is desirable in the medical domain since experts have to justify their judgments to the patients. In this work, we proposed a method for explanation-guided training that uses a layer-wise relevance propagation technique to force the model to focus only on the relevant part of the image. We experimentally verified our method on a convolutional neural network model for low-grade and high-grade glioma classification problems. Our experiments produced promising results in the way where we use interpretation techniques in the training process.

Early language exposure affects neural mechanisms of semantic representations.

One signature of the human brain is its ability to derive knowledge from language inputs, in addition to nonlinguistic sensory channels such as vision and touch. How does human language experience modulate the mechanism by which semantic knowledge is stored in the human brain? We investigated this question using a unique human model with varying amounts and qualities of early language exposure: early deaf adults who were born to hearing parents and had reduced early exposure and delayed acquisition of any natural human language (speech or sign), with early deaf adults who acquired sign language from birth as the control group that matches on nonlinguistic sensory experiences. Neural responses in a semantic judgment task with 90 written words that were familiar to both groups were measured using fMRI. The deaf group with reduced early language exposure, compared with the deaf control group, showed reduced semantic sensitivity, in both multivariate pattern (semantic structure encoding) and univariate (abstractness effect) analyses, in the left dorsal anterior temporal lobe (dATL). These results provide positive, causal evidence that language experience drives the neural semantic representation in the dATL, highlighting the roles of language in forming human neural semantic structures beyond nonverbal sensory experiences.

Construction and Application of LSTM-Based Prediction Model for Tunnel Surrounding Rock Deformation

Tunnel surrounding rock deformation is a significant issue in tunnel construction and maintenance and has garnered attention from both domestic and international scholars. Traditional methods of predicting tunnel surrounding rock deformation involve fitting monitoring and measuring data, which is a laborious and resource-intensive process with low accuracy when predicting data with significant fluctuations. A deep learning approach can improve monitoring efficiency and accuracy while reducing labor costs. In this study, taking an actual tunnel project as an example, a long short-term memory (LSTM) network model was constructed based on the recurrent neural network algorithm with deep learning to model and analyze the tunnel monitoring and measurement data, and the model was used to analyze and predict the vault settlement of the tunnel. LSTM is a type of artificial neural network architecture that is commonly used in deep learning applications for sequence prediction tasks, such as natural language processing, speech recognition, and time-series forecasting. In predicting data with smaller fluctuations, the maximum error is 4.76 mm, the minimum error is 0.03 mm, the root mean square error is 2.64, and the coefficient of determination is 0.98. In predicting data with larger fluctuations, the maximum error is 8.32 mm, the minimum error is 0.13 mm, the root mean square error is 4.42, and the coefficient of determination is 0.88. The average error of the LSTM network model is 2.16 mm. With the growth of the prediction period, the prediction results become more and more stable and closer to the actual vault settlement, which provides a reliable reference for introducing the LSTM prediction method with deep learning to tunnel construction and promoting tunnel construction safety.

A Review on Generative Adversarial Networks: Algorithms, Theory, and Applications

Generative adversarial networks (GANs) have recently become a hot research topic; however, they have been studied since 2014, and a large number of algorithms have been proposed. Nevertheless, few comprehensive studies explain the connections among different GAN variants and how they have evolved. In this paper, we attempt to provide a review of the various GAN methods from the perspectives of algorithms, theory, and applications. First, the motivations, mathematical representations, and structures of most GAN algorithms are introduced in detail, and we compare their commonalities and differences. Second, theoretical issues related to GANs are investigated. Finally, typical applications of GANs in image processing and computer vision, natural language processing, music, speech and audio, the medical field, and data science are discussed.

A survey on machine learning from few samples

The capability of learning and generalizing from very few samples successfully is a noticeable demarcation separating artificial intelligence and human intelligence. Despite the long history dated back to the early 2000s and the widespread attention in recent years with booming deep learning, few surveys for few sample learning (FSL) are available. We extensively study almost all papers of FSL spanning from the 2000s to now and provide a timely and comprehensive survey for FSL. In this survey, we review the evolution history and current progress on FSL, categorize FSL approaches into the generative model based and discriminative model based kinds in principle, and emphasize particularly on the meta learning based FSL approaches. We also summarize several recently emerging extensional topics of FSL and review their latest advances. Furthermore, we highlight the important FSL applications covering many research hotspots in computer vision, natural language processing, audio and speech, reinforcement learning and robotic, data analysis, etc. Finally, we conclude the survey with a discussion on promising trends in the hope of providing guidance and insights to follow-up researches.

Legal recognition of artificial intelligence technologies in the context of the constitutional values of the Russian state

The purpose of the article is to form approaches to the public legal regulation of artificial intelligence technologies. The subject of the work is the social relations that have developed in the field of legal registration of modern digital technologies of "weak" artificial intelligence - computer vision, natural language processing, speech recognition and synthesis, as well as intellectual decision support. The relevance of the research is determined by the need to bring the content of legal acts in line with the current level of information technology development. The work is based on a combination of general philosophical, general scientific and special methods of cognition – concrete historical analysis, formal legal (dogmatic) method, as well as the method of comparative jurisprudence. The author criticizes the anthropomorphic approach to understanding artificial intelligence, based on the analogy of technology with the human mind, implemented in the terminological apparatus of existing legal acts. It is proposed to limit the legal interpretation of artificial intelligence to a set of specific information technologies that can be used by a person in solving applied problems. The analysis of possible directions of constitutional and legal recognition of artificial intelligence technologies and the legal consequences of their influence on the implementation of the constitutional values of equality and privacy is carried out. The argumentation in favor of a combination of social and technical regulation of relations in the field of application of artificial intelligence technologies is proposed.