Research & Contribution¶

2023
- AI-Driven Webtv:An End-to-End Architecture for Automated Video Content Creation and Broadcasting
2024
- Inpage-to-Unicode Converter:A Robust Inpage-to-Unicode Encoding Converter for Kashmiri and Related Languages
- Datasets on HuggingFace
2025
- Project Title 1: AI-Based Medical Diagnosis(Comming soon)
- Project Title 2: Self-Healing IoT Networks (Comming soon)

This paper presents developing and implementing an AI-driven WebTV system designed to automate video content creation pipeline, from conceptualization to broadcasting. Leveraging open-source models such as Zeroscope for text-to-video generation and MusicGen for music synthesis, the system integrates large language models (LLMs) to create a seamless, modular production architecture.
View Paper

This paper presents a new Inpage-to-Unicode converter that achieves near 98.6% accuracy for Kashmiri by combining meticulous letter-to-letter mapping specialized diacritic handling, and an innovative chunk-based text processing architecture that keeps user interfaces responsive.
View Paper

Curated and published multiple high-quality datasets on the HuggingFace Hub for NLP , TTS, computer vision tasks.
Datasets include:
- Kashmiri Text Corpus Dataset - For NLP tasks and academic research
- Urdu Text Corpus (Sentense wise) Dataset - For NLP tasks academic research
- Persian Text Corpus Dataset - For NLP tasks academic research
- Multilingual Dictionary Dataset - For NLP (translation) tasks academic research
- 40K Kashmiri text recognition dataset - For Computer Vision and OCR Recognition tasks
- 30K Kashmiri text recognition dataset - For Computer Vision and OCR Recognition tasks
- Urdu News text Corpus Dataset - For NLP tasks academic research
- English TTS Dataset - For TTS tasks
- Scientific Papers Dataset - For academic research
All datasets are open-source and properly documented with usage examples.