Skip to main content
Myanmar text has no spaces between words — traditional spell checkers can’t handle it. mySpellChecker validates at the syllable level first, catching the vast majority of errors before expensive word segmentation even runs. 
pip install myspellchecker
mySpellChecker does not include a bundled dictionary. You must build one before spell checking will work.
from myspellchecker import SpellChecker
from myspellchecker.providers import SQLiteProvider

# Requires a built dictionary database (see Installation Guide)
provider = SQLiteProvider(database_path="mySpellChecker-default.db")
checker = SpellChecker(provider=provider)

result = checker.check("မြန်မာစာလုံးပေါင်းစစ်ဆေးခြင်း")
for error in result.errors:
    print(f"{error.text} → {error.suggestions}")

Quickstart

Install, build a dictionary, and check your first text in 5 minutes.

Try Sarsit

Use the hosted spell checker — no setup required.

​
What makes it different

Syllable-First

Validates syllable structure before word lookup. Catches 90%+ of typos at O(1) cost.

Fast Suggestions

SymSpell algorithm generates corrections 1,000x faster than edit distance search.

Context-Aware

N-gram models detect real-word errors that pass dictionary checks.

Grammar Checking

POS tagging + YAML-driven rules catch particle misuse, register mixing, and more.

AI-Powered

Optional ONNX models for semantic checking and token-level error detection.

High Performance

Cython extensions + OpenMP parallelization for production workloads.

​
Explore

Spell Checking

Syllable validation, word correction, context analysis, and grammar rules.

Dictionary Building

Build custom dictionaries from your own corpus with the data pipeline.

API Reference

Full Python API, provider capabilities, and tokenizer interfaces.