Prajesh
Bilash Panta
Crafting intelligent systems at the intersection of AI/ML research and modern web development. From voice cloning to neural image captioning — I build things that matter.
Engineering the future
I'm a Computer Engineering graduate from Kantipur Engineering College (Tribhuvan University), deeply passionate about building AI systems that solve real-world problems. My work spans neural network architectures, deep learning pipelines, and full-stack web development.
Whether it's synthesizing speech from text using FastSpeech2 or captioning images in Nepali with CNN-Transformer pipelines — I thrive at the cutting edge of technology, constantly pushing what's possible.
Things I've Built
A selection of my most significant projects, from deep learning research to production-ready systems.
Voice Cloning System
End-to-end neural voice cloning pipeline using FastSpeech2 for non-autoregressive TTS and ECAPA-TDNN for robust speaker embedding extraction. Achieves high-fidelity speech synthesis with minimal speaker samples.
Image Captioning in Nepali
Multimodal deep learning system that generates descriptive captions in the Nepali language. Combines CNN-based visual feature extraction with a Transformer decoder, bridging computer vision and NLP for low-resource languages.
Todo App
A clean and responsive full-stack Todo application featuring task creation, editing, deletion, and status tracking. Built with a focus on intuitive UX, persistent data storage, and a minimal modern interface.
Personal Portfolio
This very portfolio website — a fully custom-built, responsive personal site showcasing projects, skills, and education. Designed with smooth animations, dark/light mode, and a modern glassmorphism aesthetic using pure HTML, CSS, and JavaScript.
My Skill Set
My Education
Let's Connect
Open to opportunities
Whether you're looking for a developer to join your team, want to collaborate on an AI project, or just want to say hello — my inbox is always open.