Datasets:

sKT-Ai-Labs
/

ST-H

Name: ST-H
Creator: SKT AI LABS
License: https://choosealicense.com/licenses/apache-2.0/

Tasks:

Modalities:

Formats:

Languages:

Size:

Libraries:

License:

Dataset card Data Studio Files Files and versions

xet

Community

You need to agree to share your contact information to access this dataset

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

SKT AI LABS

SCRAPPED FROM DIFFERENT PROVIDERS

The Sovereign LLM Development For India (Project Surya)

🌌 SKT AI LABS 🌌

📂 The Hinglish-1.2G Corpus 📂

💎 Dataset Overview

Yeh 1.26 GB ka massive corpus sirf data nahi, balki modern India ki Digital Voice hai. Isme Hindi aur English ka wo flow hai jo aaj ke digital era mein sabse zyada use hota hai.

SKT AI LABS presents a high-fidelity dataset curated for premium AI development and natural language understanding.

🛠 Technical Excellence

Humne is dataset ko SKT AI LABS ke high-standards par refine kiya hai:

🔴 Massive Scale: 1.2 GB+ ka size ensure karta hai ki model har bar diverse patterns seekhe.
🟢 Natural Dialect: Real-world context jo social media aur professional discussions se inspired hai.
🔵 Clean Architecture: Manually aur automated scripts se refined taaki koi "garbage tokens" na rahein.
🟡 Instruction-Ready: Fine-tuning aur conversational training ke liye perfect format.

🚀 Vision of SKT AI LABS

Humara maqsad hai AI ko itna natural banana ki wo har Indian user ke jazbaat aur unki bhasha ko bina kisi rukawat ke samajh sake.

"Empowering Machines with the Soul of Indian Communication." > — SKT AI LABS

CONNECT US

📜 Attribution & Legal

Downloads last month: 25