GitHub – Open-source vernacular text normalization for AI pipelines

Summary: GitHub provides tools to normalize Indian code-mixed text for AI workflows, supporting LLM, RAG, and routing processes with API endpoints and Docker deployment.

What it does

It normalizes mixed-language Indian text using language packs and offers API and Docker deployment options with integration examples.

Who it's for

Teams processing real user text containing multiple scripts and languages in a single sentence.

Why it matters

It addresses challenges in handling code-mixed text for AI applications, improving text consistency before processing.