bharatlingo-ai / README.md
dpnayak251's picture
Upload README.md with huggingface_hub
56f6776 verified

A newer version of the Gradio SDK is available: 6.0.2

Upgrade
metadata
title: BharatLingo AI
emoji: ๐ŸŒ
colorFrom: green
colorTo: blue
sdk: gradio
sdk_version: 4.44.0
app_file: app.py
pinned: true
license: mit
short_description: Free AI language tools for Indian languages
tags:
  - translation
  - grammar
  - indian-languages
  - nlp

๐Ÿ‡ฎ๐Ÿ‡ณ BharatLingo AI API

Free AI-powered language tools for all Indian languages.

Setup

Required Environment Variables

DEEPSEEK_API_KEY: Your DeepSeek API key for translation fallback

  • Get your free API key from: https://platform.deepseek.com/
  • Set in HuggingFace Space Settings โ†’ Repository secrets โ†’ Add secret
  • Name: DEEPSEEK_API_KEY
  • Value: Your API key (e.g., sk-xxxxxxxxxxxxx)

Without this key, the Space will work but translation fallback will be unavailable.

Features

  • Translation: Translate between 13+ Indian languages using IndicTrans2
  • Grammar Checker: Fix grammar errors in your text
  • Spell Checker: Correct spelling mistakes
  • Text Rewriter: Rewrite text in different styles (formal, casual, professional)
  • Summarizer: Summarize long texts
  • Paraphraser: Paraphrase while keeping meaning

Supported Languages

  • English, Hindi, Bengali, Tamil, Telugu, Marathi, Gujarati
  • Kannada, Malayalam, Punjabi, Odia, Assamese, Bhojpuri

API Usage

This Space exposes API endpoints that can be called from any application:

Translation

from gradio_client import Client

client = Client("YOUR_USERNAME/bharatlingo-ai")
result = client.predict(
    text="Hello, how are you?",
    source_lang="english",
    target_lang="hindi",
    api_name="/translate"
)

Grammar Check

result = client.predict(
    text="She don't likes apples.",
    language="english",
    api_name="/grammar"
)

JavaScript/Fetch API

const response = await fetch('https://YOUR_USERNAME-bharatlingo-ai.hf.space/api/translate', {
  method: 'POST',
  headers: { 'Content-Type': 'application/json' },
  body: JSON.stringify({
    data: ["Hello world", "english", "hindi"]
  })
});
const result = await response.json();

Models Used

  • Translation: ai4bharat/indictrans2 (fallback: facebook/nllb-200)
  • Grammar: grammarly/coedit-large
  • Text Generation: google/flan-t5-base

Credits