Pub Date : 2021-11-11DOI: 10.1109/I-SMAC52330.2021.9640756
Nazrin Jariya, Kutty Malu V. K
Biometric authentication is very important and necessary. Available verification techniques contain thumbprints scanning, iris, facial and, speech recognition. Authentication using the human iris is entitled the most accurate. The human eye contains a lot of textural as well as geometrical features capable of differentiating an iris pattern separately. The iris pattern being very stable is impossible to replicate. This survey paper incorporates various methods of traditional iris recognition.
{"title":"An Overview of Advancements in Iris Recognition","authors":"Nazrin Jariya, Kutty Malu V. K","doi":"10.1109/I-SMAC52330.2021.9640756","DOIUrl":"https://doi.org/10.1109/I-SMAC52330.2021.9640756","url":null,"abstract":"Biometric authentication is very important and necessary. Available verification techniques contain thumbprints scanning, iris, facial and, speech recognition. Authentication using the human iris is entitled the most accurate. The human eye contains a lot of textural as well as geometrical features capable of differentiating an iris pattern separately. The iris pattern being very stable is impossible to replicate. This survey paper incorporates various methods of traditional iris recognition.","PeriodicalId":178783,"journal":{"name":"2021 Fifth International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132406083","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-11-11DOI: 10.1109/I-SMAC52330.2021.9640795
Arumbaka Srinivasa Rao, Yamini Tondepu, Siva Kumari N, Ch. Prasad
In the past few years, India has reported 30% of breast cancer cases, and this number is likely to increase. In India, a woman is diagnosed with breast cancer every two minutes and dies every nine minutes. Women who are diagnosed and treated early can have a better chance for survival. This article offers a new machine learning-based strategy for diagnosing breast cancer known as an Enhanced ensembled classification model. Further, this research work has conducted an experimental analysis to check the validity of the dataset extracted from the Kaggle repository. When compared to other algorithms such as Logistic Regression and SVM, the proposed model provides more accurate and effective outcomes when implemented and compared with existing methods.
{"title":"An Early Detection of Breast Cancer Using Hybrid Ensemble Classifier","authors":"Arumbaka Srinivasa Rao, Yamini Tondepu, Siva Kumari N, Ch. Prasad","doi":"10.1109/I-SMAC52330.2021.9640795","DOIUrl":"https://doi.org/10.1109/I-SMAC52330.2021.9640795","url":null,"abstract":"In the past few years, India has reported 30% of breast cancer cases, and this number is likely to increase. In India, a woman is diagnosed with breast cancer every two minutes and dies every nine minutes. Women who are diagnosed and treated early can have a better chance for survival. This article offers a new machine learning-based strategy for diagnosing breast cancer known as an Enhanced ensembled classification model. Further, this research work has conducted an experimental analysis to check the validity of the dataset extracted from the Kaggle repository. When compared to other algorithms such as Logistic Regression and SVM, the proposed model provides more accurate and effective outcomes when implemented and compared with existing methods.","PeriodicalId":178783,"journal":{"name":"2021 Fifth International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130009571","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-11-11DOI: 10.1109/I-SMAC52330.2021.9640855
K. Umamaheswari, P. Mahitha
Security is a prime aspect of concern in order to maintain confidentially of our home, work places and to avoid intrusion of unauthorized persons. In voice password and biometric based authentication door locking system, authentication using the unique identification like biometric and voice recognition plays a vital role to provide high level of security. The finger ridges of individual do not match with any other finger ridges and an individual’s voice cannot be impersonated with accuracy.This paper proposes a smart voice password and biometric based security system for door locking in smart homes. To enhance the security, in place of conventional door locking system, a finger print sensor along with a micro phone are used to authenticate, i.e. to lock and unlock the doors [1]. The data base will maintain the data of the persons who tried to access the door. The entire system is controlled by the Raspberry pi 3 B+ processor. PIWHO soft ware is used for the purpose of voice recognition. Door access may be provided to the registered users based on his voice pass word and thumb impression. Door will be opened only when both the factors are satisfied, otherwise buzzer will be activated and the authorized person will receive an SMS alert message.
{"title":"Smart security system for door access based on unique authentication","authors":"K. Umamaheswari, P. Mahitha","doi":"10.1109/I-SMAC52330.2021.9640855","DOIUrl":"https://doi.org/10.1109/I-SMAC52330.2021.9640855","url":null,"abstract":"Security is a prime aspect of concern in order to maintain confidentially of our home, work places and to avoid intrusion of unauthorized persons. In voice password and biometric based authentication door locking system, authentication using the unique identification like biometric and voice recognition plays a vital role to provide high level of security. The finger ridges of individual do not match with any other finger ridges and an individual’s voice cannot be impersonated with accuracy.This paper proposes a smart voice password and biometric based security system for door locking in smart homes. To enhance the security, in place of conventional door locking system, a finger print sensor along with a micro phone are used to authenticate, i.e. to lock and unlock the doors [1]. The data base will maintain the data of the persons who tried to access the door. The entire system is controlled by the Raspberry pi 3 B+ processor. PIWHO soft ware is used for the purpose of voice recognition. Door access may be provided to the registered users based on his voice pass word and thumb impression. Door will be opened only when both the factors are satisfied, otherwise buzzer will be activated and the authorized person will receive an SMS alert message.","PeriodicalId":178783,"journal":{"name":"2021 Fifth International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC)","volume":"99 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133995816","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-11-11DOI: 10.1109/I-SMAC52330.2021.9640718
H. Kaur
Despite of large amounts of research work on Optical Orthogonal Frequency Division Multiplexed system (OOFDM) in recent years, the area demands more exploration to investigate further potential, as it offer high spectral efficiency and flexibility. This paper simulates and presents performance analysis of signal conditioning parameters that can implement adaptivity. It reports Q-factor and bit error rate (BER) as performance metrics over varying polarization mode dispersion and chromatic dispersion for 16-QAM-OOFDM and 64-QAM-OOFDM transmissions. These signal conditioning parameters can be used for implementing adaptivity in OOFDM transmissions to achieve better transmission performance.
{"title":"Variation of Q factor over varying chromatic dispersion and polarization mode dispersion for M-QAM-OOFDM system","authors":"H. Kaur","doi":"10.1109/I-SMAC52330.2021.9640718","DOIUrl":"https://doi.org/10.1109/I-SMAC52330.2021.9640718","url":null,"abstract":"Despite of large amounts of research work on Optical Orthogonal Frequency Division Multiplexed system (OOFDM) in recent years, the area demands more exploration to investigate further potential, as it offer high spectral efficiency and flexibility. This paper simulates and presents performance analysis of signal conditioning parameters that can implement adaptivity. It reports Q-factor and bit error rate (BER) as performance metrics over varying polarization mode dispersion and chromatic dispersion for 16-QAM-OOFDM and 64-QAM-OOFDM transmissions. These signal conditioning parameters can be used for implementing adaptivity in OOFDM transmissions to achieve better transmission performance.","PeriodicalId":178783,"journal":{"name":"2021 Fifth International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC)","volume":"60 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134196740","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-11-11DOI: 10.1109/I-SMAC52330.2021.9640810
Xu Wang
With the rapid development of society, logistics companies are also presenting a rapid development model. Business managers should gradually improve their management thinking and pay attention to the importance of agile supply chain management models. Based on the supply chain management, this paper conducts an intelligent analysis of the information flow in the enterprise logistics management, and the efficiency of the results obtained by the artificial intelligence algorithm is increased by 6.3%, such as insufficient management awareness of managers, low degree of informationization, irregular business processes, and at the same time, Proposes enterprise logistics management informatization measures based on agile supply chain management, and improves the distributed node analysis of logistics information by 7%
{"title":"Analysis of Distributed Nodes of Logistics Management Information System Based on Supply Chain Management","authors":"Xu Wang","doi":"10.1109/I-SMAC52330.2021.9640810","DOIUrl":"https://doi.org/10.1109/I-SMAC52330.2021.9640810","url":null,"abstract":"With the rapid development of society, logistics companies are also presenting a rapid development model. Business managers should gradually improve their management thinking and pay attention to the importance of agile supply chain management models. Based on the supply chain management, this paper conducts an intelligent analysis of the information flow in the enterprise logistics management, and the efficiency of the results obtained by the artificial intelligence algorithm is increased by 6.3%, such as insufficient management awareness of managers, low degree of informationization, irregular business processes, and at the same time, Proposes enterprise logistics management informatization measures based on agile supply chain management, and improves the distributed node analysis of logistics information by 7%","PeriodicalId":178783,"journal":{"name":"2021 Fifth International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134198202","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-11-11DOI: 10.1109/I-SMAC52330.2021.9641050
S. T. Slevi, P. Visalakshi
The integration of IDS and Internet of Things (IoT) with deep learning plays a significant role in safety. Security has a strong role to play. Application of the IoT network decreases the time complexity and resources. In the traditional intrusion detection systems (IDS), this research work implements the cutting-edge methodologies in the IoT environment. This research is based on analysis, conception, testing and execution. Detection of intrusions can be performed by using the advanced deep learning system and multiagent. The NSL-KDD dataset is used to test the IoT system. The IoT system is used to test the IoT system. In order to detect attacks from intruders of transport layer, efficiency result rely on advanced deep learning idea. In order to increase the system performance, multi -agent algorithms could be employed to train communications agencies and to optimize the feedback training process. Advanced deep learning techniques such as CNN will be researched to boost system performance. The testing part an IoT includes data simulator which will be used to generate in continuous of research work finding with deep learning algorithms of suitable IDS in IoT network environment of current scenario without time complexity.
{"title":"A survey on Deep Learning based Intrusion Detection Systems on Internet of Things","authors":"S. T. Slevi, P. Visalakshi","doi":"10.1109/I-SMAC52330.2021.9641050","DOIUrl":"https://doi.org/10.1109/I-SMAC52330.2021.9641050","url":null,"abstract":"The integration of IDS and Internet of Things (IoT) with deep learning plays a significant role in safety. Security has a strong role to play. Application of the IoT network decreases the time complexity and resources. In the traditional intrusion detection systems (IDS), this research work implements the cutting-edge methodologies in the IoT environment. This research is based on analysis, conception, testing and execution. Detection of intrusions can be performed by using the advanced deep learning system and multiagent. The NSL-KDD dataset is used to test the IoT system. The IoT system is used to test the IoT system. In order to detect attacks from intruders of transport layer, efficiency result rely on advanced deep learning idea. In order to increase the system performance, multi -agent algorithms could be employed to train communications agencies and to optimize the feedback training process. Advanced deep learning techniques such as CNN will be researched to boost system performance. The testing part an IoT includes data simulator which will be used to generate in continuous of research work finding with deep learning algorithms of suitable IDS in IoT network environment of current scenario without time complexity.","PeriodicalId":178783,"journal":{"name":"2021 Fifth International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130751602","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-11-11DOI: 10.1109/I-SMAC52330.2021.9640653
L. Hu
With the continuous development and progress of advanced technology in the current society, the maturity of information technology has made it widely used in many industries and fields, and gradually plays an important and active role in social daily life. Based on information technology, at the same time, with the widespread application of electronic information technology, network security issues have also become a common concern. Exploring the value of electronic information technology has a positive impact on the development and progress of civil aviation. This article studies the application of information technology in the civil aviation safety management system under the background of the Internet.
{"title":"Application of Information Technology in Civil Aviation Safety Management System under the Background of Internet","authors":"L. Hu","doi":"10.1109/I-SMAC52330.2021.9640653","DOIUrl":"https://doi.org/10.1109/I-SMAC52330.2021.9640653","url":null,"abstract":"With the continuous development and progress of advanced technology in the current society, the maturity of information technology has made it widely used in many industries and fields, and gradually plays an important and active role in social daily life. Based on information technology, at the same time, with the widespread application of electronic information technology, network security issues have also become a common concern. Exploring the value of electronic information technology has a positive impact on the development and progress of civil aviation. This article studies the application of information technology in the civil aviation safety management system under the background of the Internet.","PeriodicalId":178783,"journal":{"name":"2021 Fifth International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131120372","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-11-11DOI: 10.1109/I-SMAC52330.2021.9640862
Fuyuan Mu, Zhong Wang
In view of the current situation that the electrical design and intelligent special design of some projects do not overlap each other and the docking design is out of touch, combined with the electrical intelligent design practice of a large commercial complex, the hidden dangers of electrical fire safety in heavy high-rise buildings cannot be ignored, especially In high-rise complex buildings, some high-power operating equipment in the electrical equipment will directly cause the temperature of the wires to rise, which is prone to fires. A reasonable electrical fire protection design can prevent the occurrence and spread of fires. This paper mainly studies the electrical intelligent fire safety early warning system of large-scale complex buildings and its application scenarios
{"title":"Smart Fire Safety Early Warning System of Large Complex Building based on Data Collection and Processing","authors":"Fuyuan Mu, Zhong Wang","doi":"10.1109/I-SMAC52330.2021.9640862","DOIUrl":"https://doi.org/10.1109/I-SMAC52330.2021.9640862","url":null,"abstract":"In view of the current situation that the electrical design and intelligent special design of some projects do not overlap each other and the docking design is out of touch, combined with the electrical intelligent design practice of a large commercial complex, the hidden dangers of electrical fire safety in heavy high-rise buildings cannot be ignored, especially In high-rise complex buildings, some high-power operating equipment in the electrical equipment will directly cause the temperature of the wires to rise, which is prone to fires. A reasonable electrical fire protection design can prevent the occurrence and spread of fires. This paper mainly studies the electrical intelligent fire safety early warning system of large-scale complex buildings and its application scenarios","PeriodicalId":178783,"journal":{"name":"2021 Fifth International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC)","volume":"249 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132854880","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-11-11DOI: 10.1109/I-SMAC52330.2021.9641030
Praveen Tumuluru, S. Raju, Dorababu Sudarsa, P. Rao, Sampoornamma Sudarsa, Lakshmi Burra
Nowadays, there are several instances in temples where the presence of additional monkeys irritates visitors. Not only that, but there are several times in which individuals in the villages will be harassed by money-related operations. Since tracking and splitting monkeys is a challenging task, the proposed system incorporates a camera with a built-in laser gun. When a monkey or a gathering of monkeys is discovered, the laser gun will accurately shoot the monkeys by making them unconscious for a specific amount of time. This idea will be implemented through the Internet of Things, which will identify specific entities and then notify a laser gun that is filled with unconscious injections to accurately fire such targets. The benefits of this technology include the ability to readily avoid noise and disturbance from such monkeys, as well as the ability to accurately implement computation intelligence (given by a machine learning algorithm). The modules utilized in this are a movement sensor for identifying the specially taught things in dynamic photos, a communication module that alerts to fire specific objects with accuracy, and a history module that sends the monkeys' personal information over time to the concerned center. The daily report will be delivered to the appropriate centre along with a video of monkeys being attacked with laser injections in order to render them unconscious. It could be extended in the future to detect many objects at once by using YOLOv3 or another advanced algorithm
{"title":"Smart Camera for monkeys: A Novel IoT approach for detection and controlling the monkeys using YOLOv3","authors":"Praveen Tumuluru, S. Raju, Dorababu Sudarsa, P. Rao, Sampoornamma Sudarsa, Lakshmi Burra","doi":"10.1109/I-SMAC52330.2021.9641030","DOIUrl":"https://doi.org/10.1109/I-SMAC52330.2021.9641030","url":null,"abstract":"Nowadays, there are several instances in temples where the presence of additional monkeys irritates visitors. Not only that, but there are several times in which individuals in the villages will be harassed by money-related operations. Since tracking and splitting monkeys is a challenging task, the proposed system incorporates a camera with a built-in laser gun. When a monkey or a gathering of monkeys is discovered, the laser gun will accurately shoot the monkeys by making them unconscious for a specific amount of time. This idea will be implemented through the Internet of Things, which will identify specific entities and then notify a laser gun that is filled with unconscious injections to accurately fire such targets. The benefits of this technology include the ability to readily avoid noise and disturbance from such monkeys, as well as the ability to accurately implement computation intelligence (given by a machine learning algorithm). The modules utilized in this are a movement sensor for identifying the specially taught things in dynamic photos, a communication module that alerts to fire specific objects with accuracy, and a history module that sends the monkeys' personal information over time to the concerned center. The daily report will be delivered to the appropriate centre along with a video of monkeys being attacked with laser injections in order to render them unconscious. It could be extended in the future to detect many objects at once by using YOLOv3 or another advanced algorithm","PeriodicalId":178783,"journal":{"name":"2021 Fifth International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC)","volume":"144 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133016958","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-11-11DOI: 10.1109/I-SMAC52330.2021.9640995
Asiya U A, Kiran V K
Speech emotion recognition is a very popular topic of research among researchers. This research work has implemented a deep learning-based categorization model of emotion produced by speeches based on acoustic data such as Mel Frequency Cepstral Coefficient (MFCC), chromagram, mel spectrogram etc. The developed speech emotion recognition system can recognize emotions like calm, happy, fearful, disgust, angry, neutral, surprised and sad. The Ryerson Audio-Visual Database of Emotional Speech (RAVDESS) and Toronto Emotional Speech Set (TESS) datasets were combined to enlarge our dataset which was used for speech emotion recognition. Specifically, the proposed frame work got an accuracy of 68% while using data augmentation in the RAVDESS dataset. The accuracy increased to 75% while using emotion recognition along with gender recognition in RAVDESS dataset and also by applying data augmentation techniques. Finally, the proposed framework got an accuracy of 89% while using the RAVDESS dataset and TESS datasets and various data augmentation techniques.
语音情感识别是研究人员非常关注的一个研究课题。本研究基于Mel Frequency Cepstral Coefficient (MFCC)、色谱图、Mel谱图等声学数据，实现了基于深度学习的语音情感分类模型。开发的语音情绪识别系统可以识别平静、快乐、恐惧、厌恶、愤怒、中性、惊讶、悲伤等情绪。将Ryerson情绪语音视听数据库(RAVDESS)和Toronto情绪语音集(TESS)数据集相结合，扩大我们的数据集，用于语音情绪识别。具体来说，在RAVDESS数据集中使用数据增强时，所提出的框架的准确率达到68%。在RAVDESS数据集中使用情感识别和性别识别，以及应用数据增强技术，准确率提高到75%。最后，利用RAVDESS数据集和TESS数据集以及各种数据增强技术，该框架的准确率达到89%。
{"title":"Speech Emotion Recognition-A Deep Learning Approach","authors":"Asiya U A, Kiran V K","doi":"10.1109/I-SMAC52330.2021.9640995","DOIUrl":"https://doi.org/10.1109/I-SMAC52330.2021.9640995","url":null,"abstract":"Speech emotion recognition is a very popular topic of research among researchers. This research work has implemented a deep learning-based categorization model of emotion produced by speeches based on acoustic data such as Mel Frequency Cepstral Coefficient (MFCC), chromagram, mel spectrogram etc. The developed speech emotion recognition system can recognize emotions like calm, happy, fearful, disgust, angry, neutral, surprised and sad. The Ryerson Audio-Visual Database of Emotional Speech (RAVDESS) and Toronto Emotional Speech Set (TESS) datasets were combined to enlarge our dataset which was used for speech emotion recognition. Specifically, the proposed frame work got an accuracy of 68% while using data augmentation in the RAVDESS dataset. The accuracy increased to 75% while using emotion recognition along with gender recognition in RAVDESS dataset and also by applying data augmentation techniques. Finally, the proposed framework got an accuracy of 89% while using the RAVDESS dataset and TESS datasets and various data augmentation techniques.","PeriodicalId":178783,"journal":{"name":"2021 Fifth International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133398720","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：