Create app.py

app.py

@@ -0,0 +1,58 @@
+import gradio as gr
+import pytesseract
+from PIL import Image
+from transformers import MarianMTModel, MarianTokenizer
+from nltk.tokenize import sent_tokenize
+import nltk
+
+nltk.download('punkt')
+
+# OCR function
+def ocr_image(image, language):
+    if image is None:
+        return "Please upload an image."
+    lang = '+'.join(language)
+    text = pytesseract.image_to_string(image, lang=lang)
+    return text.strip()
+
+# Translation function
+def translate_text(text, direction):
+    if not text.strip():
+        return "No text to translate."
+    src, tgt = direction.split("-")
+    model_name = f"Helsinki-NLP/opus-mt-{src}-{tgt}"
+
+    tokenizer = MarianTokenizer.from_pretrained(model_name)
+    model = MarianMTModel.from_pretrained(model_name)
+
+    sentences = sent_tokenize(text)
+    inputs = tokenizer(sentences, return_tensors="pt", padding=True, truncation=True)
+    outputs = model.generate(**inputs)
+
+    translated = [tokenizer.decode(t, skip_special_tokens=True) for t in outputs]
+    return "\n".join(translated).strip()
+
+# Gradio Interface
+iface = gr.Interface(
+    fn=ocr_image,
+    inputs=[
+        gr.Image(type="pil", label="Image"),
+        gr.CheckboxGroup(choices=["eng", "chi_sim", "fra", "deu"], value=["eng"], label="OCR Language(s)")
+    ],
+    outputs="text",
+    title="OCR Text Extractor"
+)
+
+# Add translation separately
+translate_iface = gr.Interface(
+    fn=translate_text,
+    inputs=[
+        gr.Textbox(label="Text to Translate"),
+        gr.Radio(choices=["en-zh", "zh-en", "en-fr", "fr-en"], value="en-zh", label="Translation Direction")
+    ],
+    outputs="text",
+    title="Text Translator"
+)
+
+# Combine both as a tabbed app
+gr.TabbedInterface([iface, translate_iface], ["OCR", "Translate"]).launch()