tetrak/easyocr-armenian

tetrak_hy — Armenian text recognition for EasyOCR

An Armenian text recogniser packaged as an EasyOCR custom model, trained by tetrak-hy-trainer for Tetrak, an OCR pipeline for community archives. The architecture is EasyOCR's own generation2 recognition network (VGG feature extractor, two BiLSTM layers, CTC head), so the model drops into a stock EasyOCR install.

Status: v1, alpha

These are the v1 weights, trained on synthetic line crops degraded to look like scans. v1 beats stock EasyOCR by roughly 16x on word recall and nearly doubles v0, but it does not yet beat tesseract -l hye. If you need the best available Armenian OCR today, use Tesseract's hye model. Use these weights if you are working on EasyOCR-based pipelines, or want a base to fine-tune.

Measured on real scans

Ten pages of the Armenian Soviet Encyclopedia (volume 2, pages 105-114) from Armenian Wikisource, proofread to quality level 4, with their transcriptions as ground truth. Higher is better for both figures.

Backend	Char similarity	Word recall
tesseract -l hye	0.697	0.662
marker	0.258	0.766
tesseract -l hye (auto page mode)	0.128	0.664
tetrak_hy v1	0.100	0.501
tetrak_hy v0	0.075	0.274
stock EasyOCR	0.035	0.031

Read the two columns separately. Word recall — 0.50 against Tesseract's 0.66 — is the honest measure of recognition here, and the gap it shows is real.

The char similarity column is mostly not about recognition. It is dominated by reading order on these two-column pages: Tesseract in automatic page mode reads words just as well as it does in the row above (0.664 vs 0.662 word recall) yet its char similarity collapses from 0.697 to 0.128, close to v1's, purely because the text comes out in a different order. This evaluation joins detected lines with a newline in detector order, so any backend that does not serialise two-column pages into reading order is penalised the same way. Column handling in the surrounding pipeline lifts that number without retraining anything.

Synthetic validation

98.8% word accuracy, 0.9974 normalised edit distance. Unlike v0's, the v1 validation crops are degraded exactly like the training ones, so this figure is measured on realistic input rather than clean renders — but it is still synthetic, and the real-scan table above is the one that matters.

What is next

Fine-tuning on real crops cut from scanned pages and aligned against proofread transcripts is the lever expected to close the word-recall gap. Treat v1 as a usable base, not a finished recogniser.

Files

• tetrak_hy.pth — the weights exactly as the trainer saved them (keys carry the module. prefix EasyOCR's loader expects to handle). This is the file EasyOCR loads.
• model.safetensors — the same tensors with the module. prefix stripped, for anything that isn't EasyOCR.
• tetrak_hy.yaml — charset, language list and network parameters.
• tetrak_hy.py — the architecture module EasyOCR imports by name.
• provenance.json — training recipe, dataset revision, charset and checksums for this release.

Use with EasyOCR

Download the three EasyOCR files and place them where EasyOCR looks for custom models:

from huggingface_hub import hf_hub_download

for filename in ("tetrak_hy.pth", "tetrak_hy.py", "tetrak_hy.yaml"):
    hf_hub_download("tetrak/easyocr-armenian", filename, revision="v1")

• tetrak_hy.yaml and tetrak_hy.py go in the user network directory (by default ~/.EasyOCR/user_network/).
• tetrak_hy.pth goes in the model directory (by default ~/.EasyOCR/model/).

Then:

import easyocr

reader = easyocr.Reader(["en"], recog_network="tetrak_hy")
results = reader.readtext("page.png")

Note the ["en"]: with a custom recog_network, the language list selects EasyOCR's dictionaries rather than the model — the recogniser itself is chosen by recog_network, and this model's charset covers Armenian plus basic Latin, digits and punctuation.

Pin revision= when downloading: each weights release is tagged, and provenance.json records the exact dataset revision it was trained from.

Training data

Trained on the crops-v1 configuration of tetrak/armenian-ocr-crops: 175,500 synthetic line crops of 1 to 4 consecutive tokens, rendered from proofread Armenian Soviet Encyclopedia text (Armenian Wikisource, CC BY-SA) at sizes down to 18 px and degraded with a downscale cycle, blur, tone shift, small rotation and a JPEG round-trip. Three faces were used for rendering: Noto Sans Armenian and Noto Serif Armenian (SIL Open Font Licence 1.1) and Mshtakan, which ships with macOS. No font file is redistributed here.

The exact dataset revision is in provenance.json. v0 was trained on the crops configuration — single-word crops, undegraded — which is still published for reproducibility.

Licence

The weights, like the trainer, are Apache 2.0. The training text is CC BY-SA; we publish the text itself, share-alike, in the dataset repository above, and take the position — shared by most of the ecosystem, though not legally settled — that trained weights are not a redistribution or adaptation of the training text.

Related

• tetrak-hy-trainer — synthesis, training and packaging (Apache 2.0).
• tetrak/armenian-ocr-crops — the training data (CC BY-SA 4.0).
• Tetrak — the OCR pipeline this model ships in.