---
language:
- ti
license: mit
tags:
- ocr
- trocr
- tigrinya
- ge-ez
- low-resource
- image-to-text
- transfer-learning
- transformers
- geez
- ge'ez
base_model: microsoft/trocr-base-handwritten
tasks:
- image-to-text
datasets:
- glocr
metrics:
- cer
- wer
- accuracy
model-index:
- name: tigrinya-trocr-handwritten
  results:
  - task:
      type: image-to-text
      name: Tigrinya OCR
    dataset:
      name: GLOCR Tigrinya News Text-Lines
      type: glocr
      split: test
    metrics:
    - type: cer
      value: 0.0038
      name: Character Error Rate
    - type: wer
      value: 0.0115
      name: Word Error Rate
    - type: accuracy
      value: 0.9686
      name: Exact Match Accuracy
---

# TrOCR-Handwritten for Tigrinya OCR

[![Model on HF](https://img.shields.io/badge/HuggingFace-tigrinya--trocr--handwritten-yellow?logo=huggingface)](https://huggingface.co/Yonatanhaile2026/tigrinya-trocr-handwritten)
[![License: MIT](https://img.shields.io/badge/License-MIT-blue.svg)](https://opensource.org/licenses/MIT)
[![Language: Tigrinya](https://img.shields.io/badge/Language-Tigrinya-green)]()
[![Script: Ge'ez](https://img.shields.io/badge/Script-Ge%27ez-orange)]()
[![Base: TrOCR-base-handwritten](https://img.shields.io/badge/Base-TrOCR--base--handwritten-lightgrey)](https://huggingface.co/microsoft/trocr-base-handwritten)

# Tigrinya TrOCR — Handwritten Variant

**Adapting TrOCR for Printed Tigrinya Text Recognition: Word-Aware Loss Weighting for Cross-Script Transfer Learning**

A fine-tuned [TrOCR](https://huggingface.co/microsoft/trocr-base-handwritten) model for printed Tigrinya line-level text recognition. 
This is the **handwritten pre-training variant**, fine-tuned from `microsoft/trocr-base-handwritten` using vocabulary extension and **Word-Aware Loss Weighting**
to resolve word-boundary failures caused by BPE space-marker conventions.

---

## Model Details

| Field | Value |
|---|---|
| **Model name** | `Yonatanhaile2026/tigrinya-trocr-handwritten` |
| **Base model** | `microsoft/trocr-base-handwritten` |
| **Task** | Tigrinya OCR (image-to-text) |
| **Language** | Tigrinya (`ti`) |
| **Script** | Ge'ez |
| **Model type** | VisionEncoderDecoderModel |
| **Vocabulary** | Extended from 50,265 → 50,495 tokens (230 Ge'ez characters added) |
| **Training data** | GLOCR Tigrinya News text-line images (synthetic) |

---

## Performance

Evaluated on a held-out test set of 5,000 synthetic Tigrinya text-line images.

| Metric | Value |
|---|---|
| Character Error Rate (CER) | **0.38%** |
| Word Error Rate (WER) | **1.15%** |
| Exact Match Accuracy | **96.86%** |

### Bootstrap 95% Confidence Intervals (1,000 iterations, TrOCR-Printed)

| Metric | Point Estimate | 95% CI |
|---|---|---|
| CER | 0.20% | [0.17%, 0.24%] |
| WER | 0.76% | [0.64%, 0.90%] |
| Accuracy | 97.44% | [97.02%, 97.84%] |

> Bootstrap intervals were computed on the TrOCR-Printed variant; see the printed model card for details.

### Comparison (same dataset and split)

| Model | CER | WER | Accuracy |
|---|---|---|---|
| **TrOCR-Handwritten (fine-tuned)** | **0.38%** | **1.15%** | **96.86%** |
| TrOCR-Printed (fine-tuned) | 0.22% | 0.87% | 97.20% |
| CRNN-CTC Baseline | 0.12% | 0.57% | 98.20% |

---

## Training Details

| Hyperparameter | Value |
|---|---|
| Optimizer | AdamW |
| Learning rate | `4e-5` |
| LR scheduler | Linear decay (no warmup) |
| Epochs | 10 |
| Per-device batch size | 2 |
| Gradient accumulation steps | 4 |
| Effective batch size | 8 |
| Mixed precision | FP16 |
| Boundary loss weight | 2.0 |
| Random seed | 42 |
| Training duration | ~2h 40m |
| Hardware | NVIDIA RTX 5060 Laptop (8 GB GDDR7) |

---

## How to Use

```python
from transformers import VisionEncoderDecoderModel, TrOCRProcessor
from PIL import Image

processor = TrOCRProcessor.from_pretrained("Yonatanhaile2026/tigrinya-trocrhandwritten")
model = VisionEncoderDecoderModel.from_pretrained("Yonatanhaile2026/tigrinya-trocrhandwritten")

Load your text-line image
image = Image.open("your_tigrinya_text_line.png").convert("RGB")

pixel_values = processor(images=image, return_tensors="pt").pixel_values
generated_ids = model.generate(pixel_values, num_beams=5, max_length=128)
prediction = processor.batch_decode(generated_ids, skip_special_tokens=True)[0]
print(prediction)

```
---

## Intended Use

**Suitable for:**
- Tigrinya OCR research on synthetic or clean text-line images
- Baseline comparison against printed-specific and CTC-based OCR models
- Research on cross-script transfer learning and BPE tokenizer adaptation

**Not suitable for:**
- Production OCR without validation on real scanned or handwritten documents
- Scenarios where the printed variant would be preferable
- Documents with heavy degradation, low resolution, or non-text noise

---

## Limitations

- Trained and evaluated exclusively on synthetic printed data from a single domain (newspaper text lines)
- Performance on real-world scanned or genuinely handwritten Tigrinya documents is not validated
- Underperforms the printed TrOCR variant on this synthetic printed corpus
- Results reflect a single training run on one hardware configuration

---

## Related Resources

- **Printed variant:** [`Yonatanhaile2026/tigrinya-trocr-printed`](https://huggingface.co/Yonatanhaile2026/tigrinya-trocrprinted)
- **Code repository:** [github.com/YoHa2024NKU/Tigrinya_TrOCR_Printed](https://github.com/YoHa2024NKU/Tigrinya_TrOCR_Printed)
- **Dataset:** [GLOCR — Harvard Dataverse](https://github.com/fgaim/GLOCR)

---

## Citation

If you use this model, please cite the associated paper and repository:

```bibtex
@misc{medhanie2026adaptingtrocrprintedtigrinya,
      title={Adapting TrOCR for Printed Tigrinya Text Recognition: Word-Aware Loss Weighting for Cross-Script Transfer Learning}, 
      author={Yonatan Haile Medhanie and Yuanhua Ni},
      year={2026},
      eprint={2604.20813},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2604.20813}, 
}

```
---

## License
```
MIT
```