检索结果-内蒙古大学图书馆

ieee International Conference on Acoustics, Speech, and signal processing (ICASSP)

作者： Jaitly, Navdeep Hinton, Geoffrey Univ Toronto Dept Comp Sci Toronto ON M5S 3G4 Canada

ISBN: (纸本)9781457705397

State of the art speech recognition systems rely on preprocessed speech features such as Mel cepstrum or linear predictive coding coefficients that collapse high dimensional speech sound waves into low dimensional encodings. While these have been successfully applied in speech recognition systems, such low dimensional encodings may lose some relevant information and express other information in a way that makes it difficult to use for discrimination. Higher dimensional encodings could both improve performance in recognition tasks, and also be applied to speech synthesis by better modeling the statistical structure of the sound waves. In this paper we present a novel approach for modeling speech sound waves using a Restricted Boltzmann machine (RBM) with a novel type of hidden variable and we report initial results demonstrating phoneme recognition performance better than the current state-of-the-art for methods based on Mel cepstrum coefficients.

关键词： Restricted Boltzmann machine RBM phoneme recognition TIMIT

来源：评论

学校读者我要写书评

暂无评论

Naturalistic Dialogue Management for Noisy Speech Recognition

引用

ieee JOURNAL OF SELECTED TOPICS IN signal processing 2012年第8期6卷 928-942页

作者： Passonneau, Rebecca J. Epstein, Susan L. Ligorio, Tiziana Columbia Univ Ctr Computat Learning Syst New York NY 10115 USA CUNY Hunter Coll Dept Comp Sci New York NY 10065 USA CUNY Grad Ctr New York NY 10075 USA CUNY Grad Ctr New York NY 10065 USA

With naturalistic dialogue management, a spoken dialogue system behaves as a human would under similar conditions. This paper reports on an experiment to develop naturalistic clarification strategies for noisy speech recognition in the context of spoken dialogue systems. We collected a wizard-of-Oz corpus in which human wizards with access to a rich set of clarification actions made clarification decisions online, based on human-readable versions of system data. The experiment compares an evaluation of calls to a baseline system in a library domain with calls to an enhanced version of the system. The new system has a clarification module based on the wizard data that is a decision tree constructed from three machine-learned models. It replicates the wizards' ability to ground partial understandings of noisy input and to build upon them. The enhanced system has a significantly higher rate of task completion, greater task success and improved efficiency.

关键词： Human computer interaction machine learning robustness speech system performance

来源：评论

学校读者我要写书评

暂无评论

AI-TOOLKIT: A MICROSERVICES ARCHITECTURE FOR LOW-CODE DECENTRALIZED machine INTELLIGENCE

AI-TOOLKIT: A MICROSERVICES ARCHITECTURE FOR LOW-CODE DECENT...

引用

ieee International Conference on Acoustics, Speech, and signal processing (ICASSP)

作者： Lomonaco, Vincenzo De Caro, Valerio Gallicchio, Claudio Carta, Antonio Sardianos, Christos Varlamis, Iraklis Tserpes, Konstantinos Coppola, Massimo Marmpena, Mina Politi, Sevasti Schoitsch, Erwin Bacciu, Davide Univ Pisa Pisa Italy Harokopio Univ Athens Athens Greece Informat Technol Market Leadership Athens Greece Austrian Inst Technol Seibersdorf Austria CNR Pisa Italy

ISBN: (纸本)9798350302615

Artificial Intelligence and machine learning toolkits such as Scikit-learn, PyTorch and Tensorflow provide today a solid starting point for the rapid prototyping of R&D solutions. However, they can be hardly ported to heterogeneous decentralised hardware and real-world production environments. A common practice involves outsourcing deployment solutions to scalable cloud infrastructures such as Amazon SageMaker or Microsoft Azure. In this paper, we proposed an open-source microservices-based architecture for decent-ralised machine intelligence which aims at bringing R&D and deployment functionalities closer following a low-code approach. Such an approach would guarantee flexible integration of cutting-edge functionalities while preserving complete control over the deployed solutions at negligible costs and maintenance efforts.

关键词： Artificial Intelligence Microservices Decentralized learning and Inference Pervasive Computing

来源：评论

学校读者我要写书评

暂无评论

OP-AMP SIZING BY INFERENCE OF ELEMENT VALUES USING machine learning 25

OP-AMP SIZING BY INFERENCE OF ELEMENT VALUES USING MACHINE L...

引用

International Symposium on Intelligent signal processing and Communication Systems (ISPACS)

作者： Fukuda, Masafumi Ishii, Tsukasa Takai, Nobukazu Gunma Univ Div Elect & Informat Kiryu Gunma 3768515 Japan

ISBN: (纸本)9781538621592

Along with high performance of electronic appliances, prolongation of the design period is becoming a big issue. if this problem can be solved, time spent on design can be used for circuit performance improvement and development of new circuits. Therefore, efficient circuit design through the assist of computer is required to further improve productivity. Some automatic circuit design methods have been proposed. However, these methods are unsuitable for designing a lot of circuits because it consumes a lot of time to design the new circuit. In this paper, an automatic design method of OP-Amp sizing by inference of machine learning is proposed, and predicts the element value of the circuit. From the simulation results, we succeeded in predicting element values of a circuit that satisfies the desired characteristic about 90% accuracy and shortening the design time.

关键词： machine learning Neural network analog IC automatic design regression analysis

来源：评论

学校读者我要写书评

暂无评论

Study of Convolutional Neural Network in Recognizing Static American Sign Language

Study of Convolutional Neural Network in Recognizing Static ...

引用

ieee International Conference on signal and Image processing Applications (ieee ICSIPA)

作者： Bin, Lee Yi Huann, Goh Yeh Yun, Lum Kin Tunku Abdul Rahman Univ Coll Fac Engn & Technol Dept Mech Engn Kuala Lumpur Malaysia

ISBN: (纸本)9781728133775

Sign language is a form of communication language to connect a deaf-mute person to the world. It involves the uses of hand gestures and body movement in order to express an idea. Nevertheless, general publics are mostly not educated to comprehend the sign language. For this reason, there is a need to have a translator to facilitate the communication. This paper would like to present a Convolutional Neural Network (CNN) model for predicting American Sign Language. There are 4800 images were captured to train and validate the proposed model. 95% recognition accuracy was attained in experiment, which shows robust performance in recognition 24 static American Sign Language pattern. The successful development of this model can be served as the basis to develop a more complicated sign language translator.

关键词： Convolutional Neural Network American Sign Language machine learning Computer Vision

来源：评论

学校读者我要写书评

暂无评论

Pruning Deep Neural Network Models of Guitar Distortion Effects

引用

ieee-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE processing 2023年 31卷 256-264页

作者： Sudholt, David Wright, Alec Erkut, Cumhur Valimaki, Vesa Aalborg Univ CREATE Copenhagen Denmark Queen Mary Univ London Ctr Digital Mus London England Aalto Univ Dept Signal Proc & Acoust Acoust Lab Espoo 02150 Finland

Deep neural networks have been successfully used in the task of black-box modeling of analog audio effects such as distortion. Improving the processing speed and memory requirements of the inference step is desirable to allow such models to be used on a wide range of hardware and concurrently with other software. In this paper, we propose a new application of recent advancements in neural network pruning methods to recurrent black-box models of distortion effects using a Long Short-Term Memory architecture. We compare the efficacy of the method on four different datasets;one distortion pedal and three vacuum tube amplifiers. Iterative magnitude pruning allows us to remove over 99% of parameters from some models without a loss of accuracy. We evaluate the real-time performance of the pruned models and find that a 3x-4x speedup can be achieved, compared to an unpruned baseline. We show that training a larger model and then pruning it outperforms an unpruned model of equivalent hidden size. A listening test confirms that pruning does not degrade the perceived sound quality, but may even slightly improve it. The proposed techniques can be used to design computationally efficient deep neural networks for processing the sound of the electric guitar in real time.

关键词： Integrated circuit modeling Computational modeling Real-time systems Distortion Closed box Speech processing Convolution Audio systems machine learning music supervised learning recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

DOUBLE COMPLETE D-LBP WITH EXTREME learning machine AUTO-ENCODER AND CASCADE FOREST FOR FACIAL EXPRESSION ANALYSIS 25

DOUBLE COMPLETE D-LBP WITH EXTREME LEARNING MACHINE AUTO-ENC...

引用

25th ieee International Conference on Image processing (ICIP)

作者： Shen, Fang Liu, Jing Wu, Peng Xidian Univ Sch Artificial Intelligence Xian Shaanxi Peoples R China

ISBN: (纸本)9781479970612

Although the obtained accuracy on some lab-controlled facial expression datasets has been very high, the recognition of facial expressions in wild environments is still a challenging problem. Local Binary Patterns (LBP) is a widely used operator in facial expression recognition. However, there are few variations of LBP operators specifically designed for facial expression recognition. In this paper, we propose a novel representation approach called the Double Complete d-LBP (Double Cd-LBP) according to the characteristics of facial expressions. Two d-LBP are employed to represent details and the contour of faces separately, and complete LBP is used to take sign and magnitude components into account. Moreover, multi-scale LBP is exploited to obtain local texture and global information. We then use the extreme learning machine auto-encoder (ELM-AE) as the feature selection approach to learn the discriminative feature. Cascade forest is employed as the final decision classifier. Experiments conducted on the six facial expression databases, including both lab-controlled and wild environments databases, show that our method outperforms or on par with state-of-the-arts.

关键词： Facial expression recognition local binary patterns feature extraction extreme learning machine auto-encoder cascade forest

来源：评论

学校读者我要写书评

暂无评论

Navigating and Reaching Therapeutic Goals with Dynamical Systems in Conversation-Based Interventions 48

Navigating and Reaching Therapeutic Goals with Dynamical Sys...

引用

48th ieee International Conference on Acoustics, Speech and signal processing, ICASSP 2023

作者： Ardulov, Victor Narayanan, Shrikanth University of Southern California Signal Analysis and Interpretation Lab United States

ISBN: (纸本)9781728163277

Modern human behavioral signal processing and machine-learning methods have introduced novel ways for representing and estimating internal states of people in goal-based conversational interactions, such as psychotherapy. By combining these methods with systems theoretic approaches, we demonstrate how canonical approaches to control policy design can be utilized for improving the quality of goal-oriented talk-based interactions. © 2023 ieee.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Dropout and Pruned Neural Networks for Fault Classification in Photovoltaic Arrays

引用

ieee ACCESS 2021年 9卷 120034-120042页

作者： Rao, Sunil Muniraju, Gowtham Tepedelenlioglu, Cihan Srinivasan, Devarajan Tamizhmani, Govindasamy Spanias, Andreas Arizona State Univ SenSIP Ctr Sch ECEE Tempe AZ 85281 USA Poundra LLC Tempe AZ 85281 USA Arizona State Univ Photovolta Reliabil Lab Mesa AZ 85212 USA

Automatic detection of solar array faults reduces maintenance costs and increases efficiency. In this paper, we address the problem of fault detection, localization, and classification in utility-scale photovoltaic (PV) arrays using machine learning methods. More specifically, we develop a series of customized neural networks for detection and classification of solar array faults. We evaluate fault detection and classification using metrics such as accuracy, confusion matrices, and the Risk Priority Number (RPN). We examine and assess the use of customized neural networks with dropout regularizers. We develop and evaluate neural network pruning strategies and illustrate the trade-off between fault classification model accuracy and algorithm complexity. Our approach promises to elevate the performance and robustness of PV arrays and compares favorably against existing methods.

关键词： Circuit faults Fault detection Arrays Photovoltaic systems Artificial neural networks Temperature measurement Standards Dropout neural networks machine learning photovoltaic panel fault detection pruned neural networks solar array fault classification

来源：评论

学校读者我要写书评

暂无评论

Insecurity and Hardness of Nearest Neighbor Queries over Encrypted Data 35

Insecurity and Hardness of Nearest Neighbor Queries over Enc...

引用

ieee 35th International Conference on Data Engineering (ICDE)

作者： Li, Rui Liu, Alex X. Liu, Ying Xu, Huanle Yuan, Huaqiang Dongguan Univ Technol Coll Software Engn & Cyber Secur Dongguan Peoples R China Michigan State Univ Dept Comp Sci & Engn E Lansing MI 48824 USA Hunan Univ Coll Comp Sci & Elect Engn Changsha Hunan Peoples R China

ISBN: (纸本)9781538674741

Nearest neighbor query processing is a fundamental problem that arises in many fields such as spatial databases and machine learning ASPE, which uses invertible matrices to encrypt data, is a widely adopted Secure Nearest Neighbor (SNN) query scheme. Encrypting data by matrices is actually a linear combination of the multiple dimensions of the data, which is completely consistent with the relationship between the source signals and observed signals in the signal processing. By viewing dimensions of the data and the encrypted data as source signals and observed signals, respectively, we formally prove and experimentally demonstrate that ASPE is actually insecure against even ciphertext only attacks, using signal processing theory. Prior work proved that it is impossible to construct an SNN scheme even in much relaxed standard security models, we invalidate this hardness understanding by pointing out the incorrectness of the hardness proof.

关键词： Cryptography Cloud computing Privacy Spatial databases signal processing Matrix decomposition machine learning

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：