how to prepare the data for new tessdata images in khmer lang

[mengleang-ngoun]

how to prepare the data for new tessdata images in khmer

[kba]

Khmer or Arabic?

In principle, it's the same process for every language, you need pairs of lines images and their transcription. AFAIK Khmer is a left-to-right script with distinct characters, so you basically only need ground truth data and you can train both from scratch or fine-tune the existing Khmer model.

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.