Hi,

I'am Jaydeep Chauhan

ABOUT ME

Expert in Machine Learning & AI | Industrial Sound Analysis | Python Developer | Data Analysis & Visualization

Passionate about leveraging machine learning and AI to solve complex challenges. Proven expertise in industrial sound analysis and audio event detection using Explainable AI. Skilled in Python development, data analysis, and database management. Proficient in Flask for web development and experienced with GitHub for version control. I have a Master's in Media Technology with a strong foundation in signal processing. Previously with Accenture, I led mobile application development projects and facilitated knowledge transfer. Recognized for delivering innovative solutions and awarded for outstanding contributions.

Email : 3neelkanth3@gmail.com

Place I live : Ilmenau, Germany

SKILLS

Programming Languages: Python, Golang, Javascript, C, C++

Machine & Deep Learning: Machine Learning, Deep Learning, NLP, Computer Vision, Generative AI

Data Manipulation & Analysis: Data Visualization, Data Cleaning, Feature Engineering, Feature Selection, Feature Extraction

Data Science Libraries: NumPy, Pandas, SciKit-Learn, Matplotlib, Seaborn

Databases: MS-SQL, My SQL, MongoDB, PostgreSQL

Version Control: GitHub, GitLab, SVN

Framework & Libraries: TensorFlow, Keras, PyTorch, Flask, OpenCV, PyTest

Containerization: Docker, GitLab CI/CD, GitHub Action

Other tools: HTML, CSS, Draw.io, Jira, Agile Methods, PyCharm, VSCode, Acoustic sensors

Languages: English (Fluent), German (Learning A2), Hindi (Native)

Resume

Professional Experience

Wissenschaftliche Mitarbeiter im Bereich Data Science

Jan 2025 - Present

Technische Hochschule Augsburg, Augsburg, Germany

Working on Predictive Maintenance Research: Developing predictive maintenance projects to reduce machinery downtime using Python-based machine learning models.

Preparing Research for Reputed Conferences: Conducting research and working towards presenting findings at top conferences to contribute to AI-driven industrial solutions.

Implementing AI-Driven Industrial Solutions: Exploring and applying AI techniques to enhance operational efficiency and fault detection in industrial systems.

Developing Data-Driven Insights: Analyzing industrial data to extract meaningful insights, optimize processes, and improve decision-making.

Collaborating on Cross-Functional AI Projects: Engaging with teams across engineering, data science, and operations to integrate AI-driven solutions into existing workflows.

Enhancing Predictive Analytics Frameworks: Working on improving predictive models by leveraging advanced statistical and deep learning techniques.

AI/Machine Learning R&D Engineer

Nov 2021 - Apr 2024

Fraunhofer IDMT, Ilmenau, Germany

Enhanced Engine Diagnostic Accuracy: Improved engine state classification accuracy by 10% through the development of a robust audio classification system using ResNet models. Addressed domain shift by recording and preprocessing audio data from diverse engines and locations, and generated synthetic datasets using generative AI to overcome data scarcity.

Pioneered Predictive Maintenance Solutions: Led the research in predictive maintenance, specializing in fault detection for industrial machinery. Utilized Python-based machine learning techniques to predict failures, significantly reducing potential downtime.

Advanced Manufacturing Processes: Contributed to projects in laser welding and Directed Energy Deposition (DED) arc welding, optimizing and advancing manufacturing techniques with data-driven insights.

Implemented Explainable AI Techniques: Enhanced interpretability of audio classification models by integrating Explainable AI (XAI) through sonification techniques, improving stakeholder understanding of machine learning outputs.

Managed Audio Data Pipeline: Successfully managed the entire audio data pipeline, including data synchronization, cleaning, and preprocessing, which resulted in a 20% improvement in data quality and model performance.

Increased Model Training Diversity: Augmented training data diversity by incorporating room impulse response recordings, leading to more robust and accurate audio classification models.

Innovated Synthetic Audio Generation: Spearheaded the generation of synthetic audio data using GANs, which contributed to a 15% enhancement in model generalization and performance.

Published Research in Reputed Conferences: Authored and presented research findings in top-tier conferences, contributing to the broader scientific community and solidifying expertise in AI and machine learning for industrial applications.

APPLICATION DEVELOPMENT SENIOR ANALYST (Dec 2016 - Sep 2018)

SOFTWARE ENGINEER (Dec 2015 - Nov 2016)

ASSOCIATE SOFTWARE ENGINEER (Nov 2014 - Nov 2015)

Accenture Services Pvt. Ltd., Bengaluru, India

Developed and maintained mobile applications for leading consumer goods and services clients, delivering zero-defect solutions that earned client and employer accolades.
Designed and implemented custom UIs and integrated payment gateways, enhancing user experience and transaction efficiency.
Leveraged SQL (MS-SQL, PostgreSQL, MySQL) for complex data manipulation and retrieval, ensuring optimal database performance and reliability.
Deployed applications on client servers, ensuring seamless integration and functionality in live environments.
Optimized application code for performance, scalability, and security, contributing to long-term maintainability and customer satisfaction.
Acted as a Subject Matter Expert (SME) in designing and developing mobile and web applications for multinational clients, leading to successful project outcomes.
Played a key role in Agile development, actively participating in client meetings, handling change requests, and ensuring timely delivery of projects.
Led effort estimation and served as the lead product developer for various international projects, managing full-stack development and deployment.
Maintained and set up SAT and UAT test servers, conducted unit testing, and ensured high-quality software releases.

Education

Master of Science in Media Technology

Apr 2019 - Oct 2021

Technische Universität Ilmenau, Ilmenau, Germany

Thesis: Adaptive Multi-scale Sound Event Detection

Media Project: Polyphonic Instrument Recognition in Jazz Recordings

Subjects: Digital Signal Processing, Audio Coding, Audio System Technology, Applied and Virtual Acoustics, Image, Video and Perception, Software Architecture, Deep Learning, Virtual and Augmented Reality

Final Grade: 1.8 (Good)

Bachelor of Technology

Aug 2010 - Jun 2014

G.B. Pant Engineering College, Ghurdauri, India

Major Project: The Optimized Wavelet Filters for Speech Compression

Subjects: Engineering Mathematics I & II, Fundamentals of Electronics Engineering, Basic Electrical Engineering, Electronic Devices and Circuit-1, Digital Electronics and Design Aspect, Electronics Measurement and Instrumentation, Network Analysis and Synthesis, Electromagnetic Field Theory, Analog Integrated Circuits, Signal and Systems, Digital Signal Processing, solid State Devices and Semiconductor Materials, Automatic Control Systems, Analog Communication, Antenna and Wave Propagation, Telecommunication Switching systems, Digital Communication, Optical Fiber Communication Systems, Wireless Communication System, Satellite Communication, Spread Spectrum, Optical Networking,Radar and Navigtion, Microwave Techniques, Data Communication Network, VLSI Technology, VLSI Circuit Designs, Digital System Design Using VHDL, Advanced Microprocessors, Fundamentals of Computer Programming, Concepts of Programming and OOPs, Computer Based Numerical Techniques, Computer Organization and Architecture, Datastructure using C++,

Final Grade: 2.6

Projects

Corn extrusion process parameter Classification

Acoustic insights into the corn extrusion process for enhanced quality control.

Arc welding fault classification

Empirical study on DED-Arc welding quality inspection using airborne sound analysis.

Classification of Joint Gap Formation in Laser Beam Butt Welding

Monitoring of Joint Gap Formation in Laser Beam Butt Welding Using Neural Network-Based Acoustic Emission Analysis.

Predominant Jazz Instrument Recognition : Empirical Studies on Neural Network Architectures

Predominant Jazz Instrument Recognition : Empirical Studies on Neural Network Architectures.

Adaptive Multi-scale Sound Event Detection

Adaptive Multi-scale Sound Event Detection.

Smart security camera to detect the Human and inform the user

The goal of this project is to develop an advanced security camera with email and message alert features as well as live streaming of the camera feed via a Web application. As part of the project, we maintained a log of the pictures and videos and generated and published the time-lapse video automatically..

go to GitHub

Object detector and object follower rover robot

The goal of this project is to develop a real-time object detection and object follower rover robot that uses the state-of-the-art, real-time object detection system YOLO.

Sentiment Analysis of IMDB Movie Reviews dataset

This project is based on Natural Language Processing (NLP) and aims to analyze the sentiments of a film comments database (IMDB). For this task, the CRNN model is used, and as an alternative approach, the ensemble method was also proposed.

go to GitHub

Kaplan-Meier Survival Analysis and Reconstructing Adjacency Matrices

Conducted Kaplan-Meier survival analysis on cancer patient data, calculating survival probabilities over time using the Kaplan-Meier estimator. Reconstructed adjacency matrices from the dataset representing the upper triangles of 10x10 adjacency matrices. Visualized them in a command-line interface (CLI) print format.

go to Code

Detection and classification of fluoride in the Ski surface using Machine Learning

Detection and classification of fluoride in the ski surface using Machine Learning and photo sensors data.

Publications

Predominant Jazz Instrument Recognition : Empirical Studies on Neural Network Architectures (EUSIPCO 2021) - Link

Musicological studies on jazz performance analysis commonly require a manual selection and transcription of improvised solo parts, both of which can be time-consuming. In order to expand these studies to larger corpora of jazz recordings, algorithms for automatic content analysis can accelerate these processes. In this study, we aim to detect the presence of predominant music instruments in jazz ensemble recordings. This information can guide a structural analysis in order to detect improvised solo parts. As the main contribution, we perform a comparative study on predominant automatic instrument recognition (AIR) in jazz ensembles using a taxonomy of 11 common instruments including singing voice. We compare the performance of three state-of-the-art convolutional neural networks (CNNs) including a recurrent variant and one with an attention mechanism. Our main finding is that while all networks perform comparably, the attention-based model learns the most compact feature representation as it is by orders of magnitude smaller than the other models.
Arc Welding Process Monitoring Using Neural Networks and Audio Signal Analysis (SMSI 2023) - Link

This paper investigates the potential of airborne sound analysis in the human hearing range for automatic defect classification in the arc welding process. We propose a novel sensor setup using microphones and perform several recording sessions under different process conditions. The proposed quality monitoring method using convolutional neural networks achieves 80.5% accuracy in detecting deviations in the arc welding process. This confirms the suitability of airborne analysis and leaves room for improvement in future work.
Empirical study on DED-Arc welding quality inspection using airborne sound analysis (60th ISC, 2023) - Link

This study explores the potential of audible range airborne sound emissions from Gas Metal Arc Welding (GMAW) to create an automated classification system using neural networks (NN) for weld seam quality inspection. Irregularities in GMAW process (oil presence, insufficient shielding gas) may lead to porosity imperfections in weld seams. Using Directed Energy Deposition-Arc additive manufacturing, aluminum (Al) and steel wall structures were produced with varying shielding gas flows or applying oil. Acoustic emissions (AE) generated during the welding process were captured using audible to ultrasonic range microphones. Mel spectrograms were computed from the AE data to serve as input to NN during training. The proposed model achieved notable accuracies in classifying both Al weld seams (83% binary, 68% multi-class) and steel welds (82% binary, 58% multi-class). These results demonstrate that employing audible range AE and NN in GMAW monitoring offers a viable method for lowlatency monitoring and valuable insights into improving welding quality.
Acoustic insights into the corn extrusion process for enhanced quality control (Inter-Noise 2024) - Link

In industrial extrusion processes, a solid material is pressed through a die to obtain products ofthe desired shape and dimension. Fluctuations in the process parameters have a significant impacton the product quality. In food extrusion, the expansion noise at the die can serve as an indicatorof the stability of the process. This study employs microphones to characterize the corn extrusionprocess, focusing on correlating acoustic emissions with predefined process parameters such as feed intake and water content. Experimental data from laboratory and industrial settings reveal distinctdomain shifts, yet consistent findings confirm the distinguishability of various extrusion processparameters by analyzing acoustic emissions.Employing machine learning models, includingsupport vector machines and convolutional neural networks, in conjunction with audio featuressuch as log Mel spectrograms, yields promising accuracies above 90% in discriminating betweenstandard and non-standard process parameters. The proposed acoustic quality control approachhas the potential to enhance the stability of extrusion processes and contributes to the developmentof automated monitoring systems in the field of food extrusion. This ensures consistent quality andreduced waste, ultimately leading to significant cost savings in production.

Certifications

C++ Programming - The Complete Course, Udemy, 2024 - Link
Machine Learning A-Z : AI, Python & R + ChatGPT Prize [2024], Udemy, 2024 - Link
LLMs Mastery : Complete Guide to Transformers & Generative AI, Udemy, 2024 - Link
Applications of AI for Anomaly Detection, NVIDIA, 2022 - Link
MLOps Proficiency Certificate, Abacus.AI, 2022 - Link
Complete Python Bootcamp From Zero to Hero in Python, Udemy, 2021 - Link
ARKIT & Unity 3D, creating Augmented Reality Apps with C#, Udemy, 2021 - Link

Hi, I'am Jaydeep Chauhan