Model and approach research for Tibetan spell-checking model

[tenzinyonten]

Description:
We want to build a spell-checking model to correct errors in Tibetan text sourced from OCR outputs, PDFs, and epubs. Before implementation, we'll research existing models and approaches for Tibetan (or similar low-resource language) spell-checking and OCR error correction, weigh their tradeoffs, and recommend the best one to move forward with.

Sub Task

• Research and compare existing models/approaches for Tibetan spell-checking and OCR error correction.
• Write a blog post recommending a model/approach to move forward with.

Reviewer

• @kaldan007

[lobsangshakya]

add reviewer -> @kaldan007

[Jampaa]

model Ti spell to be used

[tenzinyonten]

Blogpost: https://forum.openpecha.org/t/which-spell-checking-approach-should-we-use-for-tibetan-text/612?u=yonten Recommendation is TiSpell.