I'm Tejasva Maurya, an AI and Machine Learning Engineer passionate about developing innovative solutions and advancing the field through research and development.

• 🔭 I’m currently working on cutting-edge AI projects, focusing on visual models and natural language processing.
• 🌱 I’m deepening my knowledge in advanced machine learning techniques, including Convolutional Neural Networks (CNNs) and Generative Adversarial Networks (GANs).
• 👯 I’m seeking opportunities to collaborate on AI-driven projects that push the boundaries of technology.
• 💬 Ask me about AI, Machine Learning, Data Science, or any tech-related topics.
• 📫 How to reach me: mauryatejasva512004@gmail.com
• ⚡ Fun fact: I enjoy exploring new AI applications and watching anime.

• Languages:
• Machine Learning & AI:
• Data Science:
• Tools:
• Big Data:
• Cloud:
• Dev Tools:

• Vehicle Detection and Classification:

  • Designed a real-time vehicle classification system capable of categorizing vehicles into seven types, including buses, cars, trucks, and more. Created a custom dataset of 2100 images, manually annotated each class, and optimized the model for real-time performance.
  • Technologies Used:
    • Frameworks: TensorFlow, OpenCV
    • Skills Applied: CNNs, Image Classification, Dataset Annotation
  • Key Features:
    • Optimized for traffic monitoring and smart city applications.
    • Robust classification across diverse vehicle types.

• Speech Synthesis with SpeechT5:

  • Fine-tuned the SpeechT5 model for text-to-speech tasks using a custom dataset, including English and Hindi sentences. The dataset featured technical vocabulary and conversational sentences, making it ideal for personalized applications.
  • Technologies Used:
    • Frameworks: Hugging Face Transformers, PyTorch
    • Skills Applied: NLP, Speech Processing, Model Fine-Tuning
  • Key Features:
    • Multilingual and high-quality speech synthesis.
    • Ideal for customer support and AI-driven voice assistants.

• Optical Character Recognition (OCR) with GOT 2.0 Model:

  • Implemented the GOT 2.0 model for Optical Character Recognition (OCR) tasks, designed to extract and process textual data from images with high accuracy. The model was optimized to run on CPU, showcasing efficient resource utilization without compromising performance.
  • Technologies Used:
    • Model: GOT 2.0
    • Frameworks: PyTorch, OpenCV
    • Skills Applied: OCR Implementation, Model Optimization for CPU, Text Processing
  • Key Features:
    • Optimized the model to operate on CPU, making it feasible for resource-constrained environments.
    • Applications in document digitization, image-based text extraction, and automated data entry systems.

• Human-Computer Interaction (HCI) Using Gestures:

  • Created a real-time gesture-based human-computer interaction (HCI) system. Leveraged MediaPipe for hand tracking, offering intuitive control of devices and applications.
  • Technologies Used:
    • Frameworks: TensorFlow, MediaPipe, OpenCV
    • Skills Applied: Computer Vision, Real-Time Processing
  • Key Features:
    • Applications in accessibility, gaming, and augmented reality.
    • Efficient and low-latency implementation.

• Face Recognition and Classification:

  • Developed a face recognition and classification system using Haar Cascades for face detection and Support Vector Machines (SVM) for classification. This project provided hands-on experience with foundational facial recognition technologies and effective machine learning algorithms.
  • Technologies Used:
    • Frameworks: OpenCV, Scikit-learn
    • Skills Applied: Image Processing, Feature Extraction, SVM for Classification
  • Key Features:
    • Real-time face recognition system.
    • Robust classification with efficient detection and recognition.
    • Applications in event management, security, and media analytics.

• Sentiment Analysis on Articles:

  • Performed sentiment analysis on articles using a customized word list and machine learning models. This project demonstrated expertise in NLP and data visualization for content evaluation.
  • Technologies Used:
    • Libraries: NLTK, Scikit-learn, Matplotlib
    • Skills Applied: Sentiment Classification, Data Visualization
  • Key Features:
    • Extracts sentiment trends for business intelligence.
    • Clear and interactive dashboards for insights.

• Virtual Desktop Assistant:

  • Developed a desktop assistant that automates tasks such as opening applications, searching the web, managing files, and scheduling tasks using voice commands.
  • Technologies Used:
    • Libraries: SpeechRecognition, pyttsx3, OS, and Python’s subprocess module
    • Skills Applied: Speech-to-Text, Text-to-Speech, Task Automation
  • Key Features:
    • Natural language processing for efficient interaction.
    • Simplifies day-to-day operations through voice-enabled automation.

• AI Applications Across Various Fields: Engaged in research to explore innovative applications of AI across different industries, aiming to develop solutions that address real-world challenges and contribute to technological advancement.

• Computer Vision: Expertise in designing and deploying CNNs for real-time detection, classification, and tracking.
• Deep Learning: Proficient in advanced architectures like CNNs, GANs, Transformers, and SpeechT5.
• Natural Language Processing (NLP): Experienced in sentiment analysis, OCR, and fine-tuning large language models.
• Data Engineering: Skilled in creating custom datasets through manual annotation and automated tools.
• Tools & Frameworks: Hands-on experience with TensorFlow, PyTorch, OpenCV, Hugging Face, Scikit-learn, and MediaPipe.

• Researching and developing innovative AI solutions for various industries.
• Exploring AI applications in healthcare, education, and autonomous systems.

I am actively seeking opportunities to expand my skill set and contribute to impactful AI projects.

• Email: mauryatejasva512004@gmail.com
• LinkedIn: Tejasva Maurya
• Hugging Face: Tejasva Maurya