12,824 Hits in 6.5 sec

Grid-based approximation for voice conversion in low resource environments

Hadas Benisty, David Malah, Koby Crammer
2016 EURASIP Journal on Audio, Speech, and Music Processing  
In this paper, we propose the grid-based (GB) conversion method for such low resource environments, which is successfully trained using very few sentences (5-10).  ...  Mobile applications are characterized with low resources in terms of training data, memory footprint, and computational complexity.  ...  Acknowledgements The authors would like to thank Slava Shechtman, and the speech research group headed by Ron Hoory, at the IBM Research Labs, Haifa, Israel, for fruitful discussions.  ... 
doi:10.1186/s13636-016-0081-1 fatcat:ukgldpocivdtthgpthajziwap4

Background noise adaptive energy-efficient keywords recognition processor with reusable DNN and reconfigurable architecture

Guoqiang He, Xiaoling Ding, Minghao Zhou, Bo Liu, Li Li
2022 IEEE Access  
In LPM, DNN-shift (shift-based deep neural network) is used to achieve high recognition accuracy in a low background noise environment; in HPM, DNN-8bit (8bit weighted deep neural network) is used to achieve  ...  low power consumption in a high background noise environment.  ...  Multiplication Unit Cluster Scheme Based on Approximate Addition Approximate scheme Approximately low 8 bits Approximately low 10 bits Approximately low 12 bits Approximately low 14 bits Number of transistors  ... 
doi:10.1109/access.2022.3150354 fatcat:qzezlyz57bhznhmiwtm4xu44t4

Call Model and Test-Verification Methods for PS-LTE Core Equipment

Jae-Jeong Lee, Pyung-Koo Park, Byung-Chang Chung, Sang-Wan Kim, Ki-Dong Nam, Woo-Seop Rhee
2021 Electronics  
Conversely, public safety long-term evolution (PS-LTE) technology provides high-speed, low-latency, quick-setup, and high-security data access for multimedia-based mission-critical services.  ...  Existing technologies in disaster and safety communication networks (DSCNs) face problems such as narrow bandwidth, low transmission rate, and difficulty in securing channels.  ...  Conversely, PS-LTE technology can provide high-speed, low-latency, quick-setup, and high-security data access for multimedia-based mission-critical services.  ... 
doi:10.3390/electronics10202513 doaj:3beb3720c2e3466a8dff7c55ceb4682f fatcat:g6zl7e76ijbirgk6dgetadfrme

L3DAS22 Challenge: Learning 3D Audio Sources in a Real Office Environment [article]

Eric Guizzo, Christian Marinoni, Marco Pennese, Xinlei Ren, Xiguang Zheng, Chen Zhang, Bruno Masiero, Aurelio Uncini, Danilo Comminiello
2022 arXiv   pre-print
The L3DAS22 Challenge is aimed at encouraging the development of machine learning strategies for 3D speech enhancement and 3D sound localization and detection in office-like environments.  ...  In the end, we present and discuss the results submitted by all participants. L3DAS22 Challenge website:  ...  Neural beamforming techniques, such as the Filter and Sum Networks (FaSNet) [5] , provide state-of-the art results for Ambisonics-based SE and are usually suitable for low-latency scenarios.  ... 
arXiv:2202.10372v1 fatcat:epfsnophtjfdrafudh47rqdri4

Toward Understanding Soft Faults in High Performance Cluster Networks [chapter]

Jeffrey J. Evans, Seongbok Baik, Cynthia S. Hood, William Gropp
2003 Integrated Network Management VIII  
In addition, we identify consistent performance as an important requirement in cluster networks. Using this service requirement, we describe a comprehensive strategy for cluster fault management.  ...  Network degradations that negatively impact performance but do not result in failures often go unnoticed. In this paper, we classify such degradations as soft faults.  ...  Acknowledgments This work was supported in part by the U.S. Department of Energy, under Contract W-31-109-Eng-38 and NSF 9984811.  ... 
doi:10.1007/978-0-387-35674-7_14 fatcat:h7doiiiil5cg7goigkeycysouy

An Integrated Architecture for QoS-Enable Router and Grid-Oriented Supercomputer [chapter]

Chunqing Wu, Xuejun Yang
2005 Lecture Notes in Computer Science  
We propose a new approach to managing a rapidly evolving grid fabric and examine the aspect of Quality of Service in the Fabric layer and Application layer.  ...  As a voice conversation commences, a user's burst of activity (talk spurts) might be RANDOM with some average interval and the duration talk spurts approximate EXPONENTIAL statistics.  ...  computing ability and uniform access to resources in the heterogeneous Grid environment.  ... 
doi:10.1007/11534310_127 fatcat:qxt3xbvu6bbkdoiyirlreklcg4

Students with Low Entry Scores Succeed At University

Bernadette Lynch, Shalene Werth
2007 International Journal of Pedagogies and Learning  
of a particular, conversational kind of 'voice' in text based materials.  ...  The teaching team speculate that scaffolding facilitated the students' transition into the university as an academic milieu and thus enhanced their prospects for academic success in the course.  ...  Specifically, while no low OP students achieved a pass or higher in 2003, approximately 6% did so in 2004 and 19% did so in 2005.  ... 
doi:10.5172/ijpl.3.3.14 fatcat:qwmonu5fgvhn3j5yqfhntt5qty

Spatially explicit residential and working population assumptions for projecting and assessing natural capital and ecosystem services in Japan

Takanori Matsui, Chihiro Haga, Osamu Saito, Shizuka Hashimoto
2018 Sustainability Science  
In this study, we developed a coupling method for narrative scenarios and spatially explicit residential and working population designs for all of Japan as a common data set for ecosystem scenario analysis  ...  The residential population and the working population in primary industries were calculated using a gravity-based allocation algorithm in a manner consistent with the storylines of the PANCES scenarios  ...  Environment, Japan) and Grant-in-Aid for JSPS Research Fellow (Number 18J20266) from Japan Society for the Promotion of Science (JSPS).  ... 
doi:10.1007/s11625-018-0605-y fatcat:mm7ymq53e5ayzdmyqkfiqe523a

The personal Grid e-workspace (g-work) [chapter]

M.N. Vafopoulos, G.A. Gravvanis, A.N. Platis
2006 Grid Technologies  
Specifically, semantic web technologies and grid computing are emerging as the enabling infrastructure for the next decades because they are providing a unified way for managing distributed computing,  ...  data and human resources.  ...  An Intelligent User Interface enables the inhabitants of the AmI environment to control and interact with the environment in a natural (voice, gestures) and personalized way (preferences, context).  ... 
doi:10.2495/1-84564-055-1/08 fatcat:dkiqfu4bqzcujgbotqylfbupt4

The Current Situation, Problems and Countermeasures of China's Energy Industry Based on Five Development Concepts

Yuetong Shang, Yingying Wang
2019 DEStech Transactions on Environment Energy and Earth Science  
In the past 40 years of reform and opening up, China's energy industry has undergone tremendous changes.  ...  The total energy production and consumption has jumped to the top in the world, and the proportion of clean energy consumption has continued to increase, injecting a steady stream of momentum into social  ...  The share of natural gas in China's primary energy consumption structure has increased to approximately 7.2%.  ... 
doi:10.12783/dteees/icner2018/28536 fatcat:vwwqtyvfkrbmbpwvlhb7ioh3je

A Bayesian grid method PCA-based for mobile robots localization in unstructured environments

Joao Rodrigues, Carlos Cardeira, Fernando Carreira, Joao M. F. Calado, Paulo Oliveira
2013 2013 16th International Conference on Advanced Robotics (ICAR)  
The use of PCA in a Bayesian probabilistic grid allows to fuse the highly compressed PCA database information, obtained with the low computational effort, in an environment where repetitive scenarios can  ...  The method resorts to a PCA-based positioning sensor, filtered in a Bayesian probabilistic grid and combined with linear Kalman filters to estimate the global pose of mobile robots.  ...  In [7] , [8] , it was used for a self-localization system with a high compression ratio of the environment and low computational efforts [16] .  ... 
doi:10.1109/icar.2013.6766487 dblp:conf/icar/RodriguesCCCO13 fatcat:bjno6i6dinaevje3q2o5gm6zhu

L3DAS21 Challenge: Machine Learning for 3D Audio Signal Processing

Eric Guizzo, Riccardo F. Gramaccioni, Saeid Jamili, Christian Marinoni, Edoardo Massaro, Claudia Medaglia, Giuseppe Nachira, Leonardo Nucciarelli, Ludovica Paglialunga, Marco Pennese, Sveva Pepe, Enrico Rocchi (+2 others)
2021 2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP)  
We provide baseline models and results for both tasks, obtained with state-of-the-art architectures: FaSNet for SE and SELDnet for SELD.  ...  Usually, machine learning approaches to 3D audio tasks are based on single-perspective Ambisonics recordings or on arrays of single-capsule microphones.  ...  Neural beamforming techniques as Filter and Sum Networks (FaS-Net) [5] provide state-of-the art results for Ambisonics-based SE and are usually suitable for low-latency scenarios.  ... 
doi:10.1109/mlsp52302.2021.9596248 fatcat:uhlxd2ow35dsjidkytzz52hhnq

Wearable Sensor-Based Location-Specific Occupancy Detection in Smart Environments

Md Abdullah Al Hafiz Khan, Nirmalya Roy, H. M. Sajjad Hossain
2018 Mobile Information Systems  
Our experimental results on real-life data traces in natural settings show that our cross-modal approach can achieve approximately 0.53 error count distance for occupancy detection accuracy on average.  ...  We evaluate our approach in different contexts: conversational, silence, and mixed scenarios in the presence of 10 people.  ...  As human voice ranges approximately from 300 Hz to 4000 Hz, we lter each of the segments based on that frequency range using the band pass lter.  ... 
doi:10.1155/2018/4570182 fatcat:qixq4esooncebpc475vr323gye

Infrastructure-less Occupancy Detection and Semantic Localization in Smart Environments

Md Abdullah Al Hafiz Khan, H M Sajjad Hossain, Nirmalya Roy
2015 Proceedings of the 12th EAI International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services  
We opportunistically exploit smartphone's acoustic sensors in a conversing environment and motion sensors in absence of any conversational data.  ...  Our experimental results on real-life data traces in natural settings show that using this hybrid approach, we can achieve approximately 0.76 error count distance for occupancy detection accuracy on average  ...  As human voice ranges approximately 300 Hz to 4000 Hz, we filter each of the segments based on that frequency range using band pass filter.  ... 
doi:10.4108/eai.22-7-2015.2260062 dblp:journals/casa/KhanHR15 fatcat:65iu4spfincfvpd5ptk3nadh74

Volumetric Estimation of Coal Resources in Seam VI for Require Backfill Materials of Barapukuria Coal Mine, Dinajpur, Bangladesh

Md. Mostafijul Karim
2013 Earth Sciences  
, the only coal mine of the country under production, which capable of a low recovery with sustaining mining method.  ...  Through the distinguish software application along with formal calculation method the study represents a volumetric estimation of require backfill Materials of Barapukuria coal mine in north western Bangladesh  ...  Acknowledgement Special thanks to the ministry of Science and Technology for their financial support during the research work.  ... 
doi:10.11648/ fatcat:42chxdrvbrcnhbpev7skxa5snu
« Previous Showing results 1 — 15 out of 12,824 results