scholarly journals The Investigation of Different Loss Functions with Capsule Networks for Speech Emotion Recognition

2021 ◽  
Vol 2021 ◽  
pp. 1-12
Author(s):  
Anfernee Joan B. Ng ◽  
Kun-Hong Liu
Keyword(s):  
Emotion Recognition ◽  
Spatial Information ◽  
Image Features ◽  
Research Topic ◽  
Loss Functions ◽  
Speech Emotion Recognition ◽  
Important Research ◽  
Important Research Topic ◽  
The Common ◽  
Better Than

Speech emotion recognition (SER) is an important research topic. Image features like spectrograms are one of the common ways of extracting information from speech. In the area of image recognition, a relatively novel type of network called capsule networks has shown good and promising results. This study aims to use capsule networks to encode spatial information from spectrograms and analyse its performance when paired with different loss functions. Experiments comparing the capsule network with models from previous works show that the capsule network performs better than them.

Conformal Optimization Algorithm for Undevelopable Surfaces and its Application

2014 ◽  
Vol 602-605 ◽  
pp. 3570-3574
Author(s):  
Zhen Hua Luo ◽  
Fen Jiang
Keyword(s):  
Approximation Algorithms ◽  
Optimization Algorithm ◽  
Manufacturing Process ◽  
Research Topic ◽  
Important Research ◽  
Computer Aided Geometric Design ◽  
Industrial Manufacturing ◽  
Important Research Topic ◽  
Computer Aided ◽  
Planar Materials

In the industrial manufacturing process, most kinds of surfaces are processed by planar materials, but undevelopable surfaces are difficult develop to the plane. The approximation algorithms to develop a undevelopable surface is an important research topic in Computer Aided Geometric Design (CAGD). In this paper, we propose a new approximation algorithms based optimization algorithm. We guarantee the deformation vector make the minimum changes during the developing process. In the paper, some numerical example are given and the can illustrate the our method is effective.

Speech Emotional Features Extraction Based on Electroglottograph

2013 ◽  
Vol 25 (12) ◽  
pp. 3294-3317 ◽  
Author(s):  
Lijiang Chen ◽  
Xia Mao ◽  
Pengfei Wei ◽  
Angelo Compare
Keyword(s):  
Emotion Recognition ◽  
Speech Signal ◽  
Vocal Tract ◽  
Vocal Folds ◽  
Distribution Coefficients ◽  
Speech Emotion Recognition ◽  
Support Vector ◽  
Power Law Distribution ◽  
Transform Coefficients ◽  
Better Than

This study proposes two classes of speech emotional features extracted from electroglottography (EGG) and speech signal. The power-law distribution coefficients (PLDC) of voiced segments duration, pitch rise duration, and pitch down duration are obtained to reflect the information of vocal folds excitation. The real discrete cosine transform coefficients of the normalized spectrum of EGG and speech signal are calculated to reflect the information of vocal tract modulation. Two experiments are carried out. One is of proposed features and traditional features based on sequential forward floating search and sequential backward floating search. The other is the comparative emotion recognition based on support vector machine. The results show that proposed features are better than those commonly used in the case of speaker-independent and content-independent speech emotion recognition.

scholarly journals The happy victimizer phenomenon: Not found here

2014 ◽  
Vol 46 (1) ◽  
pp. 145-161
Author(s):  
Ana Jevtic ◽  
Jovan Miric
Keyword(s):  
Emotional Development ◽  
Positive Emotions ◽  
Negative Emotions ◽  
Age Groups ◽  
Research Topic ◽  
Important Research ◽  
Important Research Topic ◽  
Significant Difference ◽  
Moral Explanations

Children?s attribution of emotions to a moral transgressor is an important research topic in the psychology of moral and emotional development. This is especially because of the so-called Happy Victimizer Phenomenon (HVP) where younger children attribute positive emotions to a moral transgressor described in a story. In the two studies that we have conducted (children aged 5, 7 and 9, 20 of each age; 10 of each age in the second study) we have tested the possible influence of the fear of sanctions and the type of transgression (stealing and inflicting body injuries) on the attribution of emotions. Children were presented with stories that described transgressions and they were asked to answer how the transgressor felt. The fear of sanctions did not make a significant difference in attribution but the type of transgression did - more negative emotions were attributed for inflicting body injuries than for stealing. Positive emotions were explained with situational-instrumental explanations in 84% of cases while negative emotions were explained with moral explanations in 63,5%. Girls attributed more positive emotions (61%) than boys (39%). However, our main finding was that, for the aforementioned age groups, we did not find the HVP effect although it has regularly been registered in foreign studies. This finding denies the generalizability of the phenomenon and points to the significance of disciplining styles and, even more so, culture for children?s attribution of emotions to moral transgressors.

scholarly journals Facial Landmark Annotation and Fuzzy Model to Detect Players’ Focus

2021 ◽  
Author(s):  
Alisson Steffens Henrique ◽  
Esteban Walter Gonzalez Clua ◽  
Rodrigo Lyra ◽  
Anita Maria da Rocha Fernandes ◽  
Rudimar Luis Scaranto Dazzi
Keyword(s):  
Fuzzy Model ◽  
Research Topic ◽  
Important Research ◽  
Facial Landmark ◽  
Important Research Topic ◽  
Focus Detection ◽  
Data Log ◽  
Game Analytics

Game Analytics is an important research topic in digitalentertainment. Data log is usually the key to understand players’behavior in a game. However, alpha and beta builds may need aspecial attention to player focus and immersion. In this paper, wepropose t he us e of player’s focus detection, through theclassification of pictures. Results show that pictures can be usedas a new source of data for Game Analytics, feeding developerswith a better understanding of players enjoyment while in testingphases .

Structuring the Cultural Domain with an Upper Ontology of Culture

2011 ◽  
pp. 179-212 ◽  
Author(s):  
Emmanuel G. Blanchard ◽  
Riichiro Mizoguchi ◽  
Susanne P. Lajoie
Keyword(s):  
Neutral Theory ◽  
Important Research ◽  
Communication Management ◽  
Automatic Reasoning ◽  
Important Research Topic ◽  
Cultural Domain ◽  
The Common ◽  
Similarities And Differences ◽  
Technology Specialists ◽  
Upper Ontology

Study of cultural similarities and differences is an important research topic for many disciplines such as psychology, sociology, anthropology, archaeology, museology, communication, management and business. This presents many potential opportunities for Information Technology specialists to develop culturally-aware technology, but it also raises the risk of inconsistent approaches of the cultural domain. In this chapter, the authors present the fundamental concepts of the Upper Ontology of Culture (UOC), a formal conceptualization of the cultural domain they developed by identifying the common backbone of culture-related disciplines and activities. As a neutral, theory-driven, and interdisciplinary conceptualization, the UOC shall provide guidelines for the development of culturally-aware applications, for the consistent computerization of cultural data and their interoperability, as well as for the development of culture-driven automatic reasoning processes.

Trends and Transformations in European Agricultural Economy, Rural Communities and Food Sustainability in Context of New Common Agricultural Policy (CAP) Reforms

2016 ◽  
pp. 1-24
Author(s):  
Andrei Jean-Vasile ◽  
Alexandra Alecu
Keyword(s):  
Rural Communities ◽  
Agricultural Policy ◽  
Common Agricultural Policy ◽  
Research Topic ◽  
Economic Sector ◽  
Important Research ◽  
European Level ◽  
Agricultural Economy ◽  
Security Assurance ◽  
Important Research Topic

Agriculture continues to be quite a debate for the last two and a half decades at least at the European level and for Romania Common Agricultural Policy (CAP) reforms has a big impact in developing the convergence to the European agricultural model. Agriculture becomes nowadays a multirole economic sector, with major implications on rural community's sustainability and on food security assurance. In this context, the transformations in European agricultural economy, rural communities and food sustainability in context of Common Agricultural Policy (CAP) reforms represent an important research topic in the context of EU-28 policy diversification from the larger context of Romanian approach.

Deriving Information System Security and Privacy From Value Cocreation Theory

2019 ◽  
Vol 10 (4) ◽  
pp. 1-25
Author(s):  
Christophe Feltus
Keyword(s):  
Information System ◽  
Value Chain ◽  
System Architecture ◽  
Enterprise Architecture ◽  
Research Topic ◽  
Security And Privacy ◽  
Important Research ◽  
Important Research Topic ◽  
Value Cocreation ◽  
The Relationship

Traditionally, the relationship between the company and its providers have for objective to generate value at the company side in exchange of money. This relationship is largely investigated through the vector of value chain. In this article, security and privacy cocreation (SPCC) is investigated as a specialization of value cocreation. Although it is an important research topic, and despite a plethora of research aiming at depicting the fundamental of SPCC, few contributions have been appeared until now in the area of a language to support SPCC design and deployment. However, such a language is necessary to describe elements of the information system, as well as their underlying dependencies. As a result, this article proposes extending an existing enterprise architecture language to support the process of decision-making and to allow understanding and analysis of the impacts associated to a change of the system architecture as a whole.

scholarly journals Vertex-Disjoint Paths in a 3-Ary n-Cube with Faulty Vertices

2020 ◽  
Vol 2020 ◽  
pp. 1-14
Author(s):  
Xiaolei Ma ◽  
Shiying Wang
Keyword(s):  
Interconnection Networks ◽  
Transmission Rate ◽  
Research Topic ◽  
Disjoint Paths ◽  
Important Research ◽  
Important Research Topic ◽  
Vertex Disjoint

The construction of vertex-disjoint paths (disjoint paths) is an important research topic in various kinds of interconnection networks, which can improve the transmission rate and reliability. The k-ary n-cube is a family of popular networks. In this paper, we determine that there are m2≤m≤n disjoint paths in 3-ary n-cube covering Qn3−F from S to T (many-to-many) with F≤2n−2m and from s to T (one-to-many) with F≤2n−m−1 where s is in a fault-free cycle of length three.

Nanomaterials and Nanocomposites for Adsorptive Desulfurization

2021 ◽  
pp. 1629-1658
Author(s):  
Saddam A. Al-Hammadi
Keyword(s):  
Research Topic ◽  
Low Energy ◽  
Important Research ◽  
Operation Condition ◽  
Adsorptive Desulfurization ◽  
Important Research Topic ◽  
Low Energy Consumption ◽  
Sulfur Adsorption ◽  
Sulfur Containing ◽  
Sulfur Containing Compounds

Desulfurization (removal of S compounds) of fuels is an important research topic in recent years. Several techniques have been reported to remove the sulfur-containing compounds in fuels. One of these techniques is adsorptive desulfurization (removal based on chemisorption and physisorption), which has received much attention because of low energy consumption and facile operation condition. This chapter discusses the methods employed under this technique and the types of nanocomposites and hybrid materials (adsorbents) that have been investigated as potential adsorbents. The strategies to enhance sulfur adsorption capacity and main challenges will be discussed.

Connectivity and Nature Diagnosability of Leaf-Sort Graphs

2020 ◽  
Vol 20 (03) ◽  
pp. 2050011
Author(s):  
JUTAO ZHAO ◽  
SHIYING WANG
Keyword(s):  
Interconnection Networks ◽  
Interconnection Network ◽  
Research Topic ◽  
Multiprocessor System ◽  
Important Research ◽  
Edge Connectivity ◽  
Topology Structure ◽  
Pmc Model ◽  
Important Research Topic

The connectivity and diagnosability of a multiprocessor system or an interconnection network is an important research topic. The system and interconnection network has a underlying topology, which usually presented by a graph. As a famous topology structure of interconnection networks, the n-dimensional leaf-sort graph CFn has many good properties. In this paper, we prove that (a) the restricted edge connectivity of CFn (n ≥ 3) is 3n − 5 for odd n and 3n − 6 for even n; (b) CFn (n ≥ 5) is super restricted edge-connected; (c) the nature diagnosability of CFn (n ≥ 4) under the PMC model is 3n − 4 for odd n and 3n − 5 for even n; (d) the nature diagnosability of CFn (n ≥ 5) under the MM* model is 3n − 4 for odd n and 3n − 5 for even n.

Sign in / Sign up

Export Citation Format

Share Document