Design and implementation of speech recognition system integrated with internet of things

Abdulkareem, Ademola; Somefun, Tobilola Emmanuel; Chinedum, Oji K.; Agbetuyi, Ayoade F.

doi:10.11591/ijece.v11i2.pp1796-1803

Cited by 10 publications

(5 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The proposed data-poisoning 4 technique Figures 5 and 6 are based on the creation of a function that creates dynamic triggers by inserting sounds into clean audio data. Imperceptibility checks were included in this dynamic audio data poisoning to ensure that the triggers created could not be distinguished by human listeners.…”

Section: E Backdoor Attack Performance and Impact Of Dirtyflippingmentioning

confidence: 99%

“…Indeed, machine learning techniques have become a common tool in the lives of the general public, with one of the most important being automatic speech recognition. In particular, intelligent speech recognition can enable communication with systems lacking traditional interfaces, such as the Internet of Things [4] (IoT) family of devices, and increase the communication efficiency for complex systems such as autonomous driving [5]. Intelligent speech recognition has also been used recently in telecommunications for automatic surveillance [6], remote identification [7], etc.…”

mentioning

confidence: 99%

See 1 more Smart Citation

A Backdoor Approach with Inverted Labels Using Dirty Label-Flipping Attacks

Mengara

2024

IEEE Access

View full text Add to dashboard Cite

Audio-based machine learning systems frequently use public or third-party data, which might be inaccurate. This exposes deep neural network (DNN) models trained on such data to potential data poisoning attacks. In this type of assault, attackers can train the DNN model using poisoned data, potentially degrading its performance. Another type of data poisoning attack that is extremely relevant to our investigation is label flipping, in which the attacker manipulates the labels for a subset of data. It has been demonstrated that these assaults may drastically reduce system performance, even for attackers with minimal abilities. In this study, we propose a backdoor attack named "DirtyFlipping", which uses dirty label techniques, 'label-on-label', to input triggers (clapping) in the selected data patterns associated with the target class, thereby enabling a stealthy backdoor.INDEX TERMS Poisoning attacks, Backdoor attacks, Adversarial machine learning.

show abstract

Section: E Backdoor Attack Performance and Impact Of Dirtyflippingmentioning

confidence: 99%

mentioning

confidence: 99%

A Backdoor Approach with Inverted Labels Using Dirty Label-Flipping Attacks

Mengara

2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…IMF can characterize each frequency component of structural response signals, having independence characteristics between components. Any signal processed by EMD can be expressed as the sum of IMFs and the residual, which shows great advantages in processing nonstationary signals, and has been widely used in voice signal recognition, mechanical fault diagnosis, SDI and other fields [42][43][44][45].…”

Section: Imf Characteristicsmentioning

confidence: 99%

A new convolutional neural network-based framework and data construction method for structural damage identification considering sensor placement

Yang

Peng

2023

Meas. Sci. Technol.

View full text Add to dashboard Cite

In the application of data driven Structural Damage Identification (SDI) based on supervised deep learning technology, valid data demarcation is the foundation; a convolutional neural network model with learning ability and capability of processing rich signal information is the core. Based on this understanding, this work makes three contributions: Firstly, the structural damage location and severity are jointly demarcated, and the SDI problem is transformed into a multi-classification task. Secondly, a 3D Signal processing Convolutional Neural Networks (3DS-CNN) is designed with an attempt to identify the complex and slight damages using the most basic network structure. Thirdly, a “Major and Subsidiary” Data Construction (MSDC) method integrating the key Intrinsic Mode Function (IMF) is proposed to construct 3D data. Then the proposed schemes are verified by two different structures. The results show that the 3DS-CNN has excellent damage identification ability for small-size data with noise pollution. MSDC method can enrich the feature information of the damage signals and help the network with deep feature excavation, even if the vibration signals are heavily polluted. Going one step further, the impact of sensor placement is discussed, and it is found that when external excitation is obvious, better SDI accuracy can be achieved even using a single sensor signal with slight noise. When the noise interference is obvious, the generalization ability and noise robustness of the network can be enhanced by optimizing sensor placement. In this case, the sensor placement criteria and the sensitive nodes of the structure should be comprehensively and carefully considered to avoid mutual “coupling” interference of data between sensors.

show abstract

“…Django adopts MTV architecture mode (i.e., model M, view v, and template T), which is an open source code of web application architecture written in Python [18]. It is a better web framework in Python, integrates many modules, and makes the connection between modules more stable.…”

Section: Django (Python Web Framework) Technologymentioning

confidence: 99%

Design and Implementation of Internet of Things Technology in College Students’ Innovation and Entrepreneurship Integration System

Sun

2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

In order to improve the operation efficiency of the innovation and entrepreneurship integration system of college students, this document proposes to formulate and implement the innovation and entrepreneurship integration system of college students based on “Internet technology.” The web front end is implemented by Django technology, the back end is developed by Apache server pycharm, the database is written in Python, and the database is MySQL. The system meets the expected objectives of implementing, monitoring, and managing innovation and entrepreneurship projects. How the system realizes the functions of repeated comparison projects, easy tracking projects, and fast search, and how managers can make progress at any time. By creating different account function permissions, the system realizes the login, declaration, management, and other functions required by the users of the innovation and entrepreneurship project management system. It is very suitable for college students to improve the project application, information review, and application process. At the same time, it is helpful to supervise and select teachers and greatly improve management efficiency. The test results show that the minimum values of 30, 80, and 150 simulated users in the system performance test are 4, 5, and 7, respectively. The maximum users of the system performance test model are 30, 80, 150, 64, 284, and 398. The system performance tests are taken under the conditions of 30, 80, and 150 user bandwidth, with the results of 113/min, 104/min, and 90/min. The system basically meets the daily work management standards and fully achieves the expected system design goals. The development and use of this system are related to the development of college students’ innovation and entrepreneurship project management and have a clear reference for the development of innovation and entrepreneurship project management systems similar to manage workflow in the future.

show abstract

Design and implementation of speech recognition system integrated with internet of things

Cited by 10 publications

References 19 publications

A Backdoor Approach with Inverted Labels Using Dirty Label-Flipping Attacks

A Backdoor Approach with Inverted Labels Using Dirty Label-Flipping Attacks

A new convolutional neural network-based framework and data construction method for structural damage identification considering sensor placement

Design and Implementation of Internet of Things Technology in College Students’ Innovation and Entrepreneurship Integration System

Contact Info

Product

Resources

About