Create EADME.md

EADME.md

@@ -0,0 +1,77 @@
+---
+license: apache-2.0
+datasets:
+- FatimahEmadEldin/SaudiSpell-Error-Injection
+language:
+- ar
+metrics:
+- wer
+base_model:
+- UBC-NLP/AraT5-base
+tags:
+- auto-correction
+---
+
+
+
+# SaudiSpell-AraT5: Balanced Saudi Dialect Spelling Correction
+
+**SaudiSpell-AraT5** is a sequence-to-sequence spelling correction model fine-tuned on a massive, balanced corpus of Saudi Dialectal Arabic. Unlike generic Arabic correctors, this model is specifically engineered to handle the nuances of **Najdi**, **Hijazi**, and **Standard Saudi** dialects alongside Modern Standard Arabic (MSA).
+
+## Model Description
+
+- **Base Model:** [UBC-NLP/AraT5-base](https://huggingface.co/UBC-NLP/AraT5-base)
+- **Task:** Grammatical Error Correction (GEC) / Spelling Correction
+- **Dialect Coverage:**
+    - 🛡️ **Najdi** (33% of training data)
+    - 🕋 **Hijazi** (33% of training data)
+    - 🇸🇦 **Standard Saudi** (33% of training data)
+
+## Training Methodology
+
+The model was trained on **3 Million sentences** with a strict **Balanced Sampling Strategy** to prevent the majority dialect from dominating. Rare dialects were oversampled 50x to match the volume of web-crawled data.
+
+### Synthetic Error Injection
+To mimic real-world noisy text, we used a "Saudi-Calibrated" error injection framework. The training data consists of **20%** of each of the following error types:
+1. **⌨️ Keyboard Typos:** Adjacency errors based on the Arabic keyboard layout.
+2. **🗣️ Phonetic Substitutions:** Common dialectal sound-alike errors (e.g., Qaf variations).
+3. **👁️ Visual Substitutions:** Look-alike character errors.
+4. **❌ Deletions & Insertions:** Random character noise mimicking fast typing.
+5. **✂️ Space Stripping:** Merged words mimicking ASR artifacts or missing spaces.
+
+## Usage
+
+You can use this model with the Hugging Face `transformers` library.
+
+```python
+from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
+
+model_name = "{FULL_REPO_ID}"
+
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
+
+def correct_text(text):
+    # Note: The model was trained with the prefix "correct: "
+    input_text = "correct: " + text
+    
+    inputs = tokenizer(input_text, return_tensors="pt", max_length=128, truncation=True)
+    
+    outputs = model.generate(
+        **inputs, 
+        max_new_tokens=128,
+        num_beams=5,        # Beam search for better quality
+        early_stopping=True
+    )
+    return tokenizer.decode(outputs[0], skip_special_tokens=True)
+
+# Example (Hijazi/Najdi Mix)
+noisy_text = "kif 7alak ya 9adi8i" # Arabizi/Typos support
+print(correct_text(noisy_text))
+
+```
+
+## Limitations
+
+* The model is optimized for short-to-medium length sentences (social media, chat).
+* While robust to some Arabizi (Latin script Arabic), it performs best on Arabic script input.