A5041793000- Speech Recognition and Synthesis
- Speech and Audio Processing
- Music and Audio Processing
- Digital and Cyber Forensics
- Advanced Malware Detection Techniques
- Security and Verification in Computing
- Web Application Security Vulnerabilities
- Big Data Technologies and Applications
- Cloud Data Security Solutions
- Digital Media Forensic Detection
- IoT and Edge/Fog Computing
- Chaos-based Image/Signal Encryption
- Machine Learning and ELM
- Information and Cyber Security
- Advanced Data Compression Techniques
- Cryptographic Implementations and Security
- Artificial Intelligence in Healthcare
- Advanced Neural Network Applications
- Big Data and Business Intelligence
- Blockchain Technology Applications and Security
- Sentiment Analysis and Opinion Mining
- Data Quality and Management
- Network Security and Intrusion Detection
- Emotion and Mood Recognition
- Electronic Health Records Systems
Netaji Subhas University of Technology
2021-2024
Institute of Food Science & Technology
2023
Guru Gobind Singh Indraprastha University
2019-2022
Indira Gandhi Delhi Technical University for Women
2020
Ambedkar University Delhi
2017
Abstract The Internet of Things (IoT) is a network embedded devices that are uniquely identifiable and have software required to communicate between the transient states. purpose this study explore discrete IoT security challenges pertaining currently deployed standards protocols. We presented detailed review in focuses on IoT's imminent aspects, covering identification risks current system, novel protocols, projects proffered recent years. This work presents an updated architecture...
A phoneme represents the smallest distinct sound unit within a language, unique to each linguistic system. Phoneme identification finds applications in various speech-related technologies, including automatic speech recognition, language learning, assistive technology for impaired individuals. The realm of Automatic Speech Recognition (ASR) stands as formidable challenge field, with predominant emphasis on extensively spoken languages. However, intricacies posed by Hindi recognition have...
Background: Speech Recognition is the most effective and suitable way of communication. Extracted features play an important role in speech recognition. Previous research works for Hindi recognition lack detailed comparative analysis feature extraction methods using dynamic energy parameters. Objective: The work presents experimental done to explore effects integrating coefficients parameters with different techniques on Connected word As extracted a significant recognition, presented show...
In this paper, one of the acoustic units speech, syllable, is used for development a continuous Hindi speech recognition system. The syllable larger unit that overcomes contextual effects and requires fewer training samples in comparison to triphone based word-based models. Other such as phoneme-based suffer from influences, context-dependent triphones due non-availability patterns with large memory storage numerous Earlier research works related were performed using word, phoneme, authors...
This paper presents a comparative analysis of acquisition methods in digital forensic. The author compares the data and memory with help Forensic toolkit imager (FTK imager) ProDiscover tool is open source forensic tools. By taking some perimeter for comparison that generally related to acquisition. In this paper, 32GB Pen drive made both tools, result taken captured, which also presented screenshot, physical (4GB RAM) made, captured screenshot paper. comparing these two gets one better than...
Brain tumor detection in MRI scans has developed into a growing study field medical imaging systems during the past few decades. Finding brain helps to pinpoint tumor’s exact size and location. This uses image augmentation convolution neural networks detect tumors. On Br35H dataset, Experiment’s trials are performed. The proposed model makes use of regularization techniques lessen overfitting. With an accuracy 98.84%, BTD-CNN exceeds all other cutting-edge methods.
Electronic Patient records (EPR) are very sensitive to individuals, so it becomes important manage the security with preciseness. Recent attacks show that improper management EPR causes a serious violation of HIPPA by seeking Protected Health Information. Managing is time and cost constraint, we need cost-efficient methods for securing records. have attributes various levels security, encrypt whole same encryption algorithm wastage power. To overcome this issue, proposed framework first...