@pakhad/locale-in
v1.0.0
Published
Indian language locale packs for pakhad — Hindi, Marathi, Tamil, Telugu, Kannada, Malayalam, Bengali, Gujarati, Punjabi
Maintainers
Readme
@pakhad/locale-in
Indian language locale packs for pakhad. Ships pre-built romanized (Latin-script) Markov models and name lists for 9 Indian languages.
Languages
| Locale | Language | Names | Export |
|--------|----------|-------|--------|
| in-hi | Hindi | 249 | hi |
| in-mr | Marathi | 164 | mr |
| in-ta | Tamil | 96 | ta |
| in-te | Telugu | 37 | te |
| in-kn | Kannada | 31 | kn |
| in-ml | Malayalam | 50 | ml |
| in-bn | Bengali | 83 | bn |
| in-gu | Gujarati | 65 | gu |
| in-pa | Punjabi | 62 | pa |
All models use romanized Latin-script names (e.g. "rahul" not "राहुल").
Install
npm install @pakhad/core @pakhad/locale-inUsually paired with @pakhad/locale-en for combined coverage:
npm install @pakhad/core @pakhad/locale-en @pakhad/locale-inUsage
All Indian locales (recommended)
import { create } from '@pakhad/core';
import en from '@pakhad/locale-en';
import indiaLocales from '@pakhad/locale-in';
const detector = create({ locales: [en, ...indiaLocales] });
detector.detect('Rahul Sharma', { fieldType: 'name' });
// { label: "clean", score: 0.086 }
detector.detect('xkqzvb wwwxxx', { fieldType: 'name' });
// { label: "gibberish", score: 0.777 }Individual locales
import { hi, mr, ta } from '@pakhad/locale-in';
// Only Hindi and Marathi
const detector = create({ locales: [en, hi, mr] });Force a specific locale
detector.detect('rahul sharma', { locale: 'in-hi' });
detector.detect('sachin patil', { locale: 'in-mr' });
detector.detect('rajesh kumar', { locale: ['in-hi', 'in-ta'] });Available Exports
// Default export: array of all 9 locales
import indiaLocales from '@pakhad/locale-in';
// Named exports: individual locales
import { hi, mr, ta, te, kn, ml, bn, gu, pa } from '@pakhad/locale-in';Expanding the Corpus
The shipped name lists are curated starter sets. For production with larger corpora, use @pakhad/train:
npx pakhad-train markov --input your-hindi-names.txt --order 3 --output in-hi-markov.bin
npx pakhad-train namelist --input your-hindi-names.txt --fpr 0.001 --output in-hi-names.bloomFull Documentation
github.com/nikhilchintawar/pakhad
License
MIT
