Terminal AI agent for Northeast India's low-resource languages. Local, offline, multilingual routing (search, translate, transcribe).
# NE-Agent
**A terminal-based AI agent for Northeast India's indigenous languages.**
*Northeast India is home to 220+ indigenous languages, yet open-source AI agents built specifically for them remain rare.*
Built on the **NE-Stack** by MWire Labs, Shillong, Meghalaya.
---
*Demo: Language detection → Tool routing → Retrieval, Translation or ASR → Local response.*
---
## Installation
```bash
pip install ne-agent
ollama pull qwen2.5:1.5b
ollama serve
ne-agent
```
Requires Python 3.9+ and Ollama. GPU recommended for translate/transcribe; search runs fine on CPU.
---
## What is NE-Agent?
NE-Agent is one of the first publicly available open-source AI agents built on indigenous Northeast Indian language infrastructure. It detects the input language, routes each query to the right tool (retrieval-augmented search, Khasi-English translation, or speech transcription), and generates grounded responses. Everything runs **locally**: no API keys, no cloud, no internet required at inference time.
Designed as a reference implementation for low-resource, multilingual agentic AI systems
---
## Features
- ✅ Fully local, no API keys, no cloud
- ✅ Offline after setup
- ✅ Retrieval-augmented generation (RAG)
- ✅ Speech transcription (ASR), 8 languages
- ✅ Khasi → English translation
- ✅ Automatic language identification (11 languages)
- ✅ LLM-based tool auto-routing
- ✅ Rich terminal UI
---
## How it works
```mermaid
flowchart TD
A[User Query text or audio] --> B[NE-LID Language Detection]
B --> C[Tool Router qwen2.5:1.5b]
C --> D[search NE-Embed + FAISS]
C --> E[translate NLLB Khasi]
C --> F[transcribe NE-ASR]
D --> G[Rich Terminal UI]
E --> G
F --> G
```
A single entry point handles heterogeneous requests, no need to specify a mode. Audio bypasses the router and goes straight to transcription; text queries are routed dynamically.
---
## Architecture
| Component | Role | Model | Notes |
|---|---|---|---|
| **NE-LID** | Language ID | fastText, 11 language …