Upload 3 files

README.md

@@ -1,12 +1,59 @@
 ---
-title: Punctuation Capitalization Correction
-emoji: 🐨
-colorFrom: pink
-colorTo: yellow
+title: Multilingual Punctuation Capitalization Correction
+emoji: 🌍
+colorFrom: blue
+colorTo: purple
 sdk: gradio
-sdk_version: 5.46.0
+sdk_version: 4.44.0
 app_file: app.py
 pinned: false
+license: apache-2.0
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# 🌍 Multilingual Punctuation & Capitalization Correction
+
+This Space provides an interactive interface for restoring punctuation, fixing capitalization, and detecting sentence boundaries in text across **47 languages**.
+
+## Features
+
+- **Multi-language support**: Works with 47 languages including English, French, Spanish, German, Italian, Portuguese, Russian, Turkish, Chinese, Japanese, Arabic, and more
+- **Three correction modes**:
+  - 📝 **Conservative**: Minimal changes, preserves original flow
+  - 📖 **With Sentence Boundaries**: Splits text into clear sentences
+  - ⚖️ **Balanced**: Smart chunking for longer texts
+- **Interactive UI**: Compare different correction styles and select the best one
+- **Copy functionality**: Easy clipboard access for each version
+
+## Model
+
+This application uses the [1-800-BAD-CODE/xlm-roberta_punctuation_fullstop_truecase](https://huggingface.co/1-800-BAD-CODE/xlm-roberta_punctuation_fullstop_truecase) model, which is an XLM-RoBERTa model fine-tuned for:
+- Punctuation restoration
+- True-casing (capitalization)
+- Sentence boundary detection
+
+## Usage
+
+1. Enter text without proper punctuation or capitalization
+2. Click "Add Punctuation & Capitalization"
+3. Review the three different correction styles
+4. Select and copy the version that best fits your needs
+
+## Examples
+
+Try these example inputs:
+- English: "hello there how are you doing today i hope everything is going well"
+- French: "bonjour comment allez vous aujourdhui jespere que tout va bien"
+- Spanish: "hola como estas espero que todo este bien contigo y tu familia"
+
+## Technical Details
+
+- **Base Model**: XLM-RoBERTa
+- **Languages Supported**: 47
+- **Tasks**: Punctuation restoration, capitalization, sentence boundary detection
+- **Framework**: Gradio interface with ONNX runtime for efficient inference
+
+## Limitations
+
+- Model was primarily trained on news data
+- May not perform optimally on conversational or informal text
+- Some languages may have better performance than others based on training data distribution

app.py

@@ -0,0 +1,215 @@
+import gradio as gr
+from punctuators.models import PunctCapSegModelONNX
+
+# Load the punctuation model
+print("Loading XLM-RoBERTa punctuation model...")
+model = PunctCapSegModelONNX.from_pretrained(
+    "1-800-BAD-CODE/xlm-roberta_punctuation_fullstop_truecase"
+)
+print("Model loaded successfully!")
+
+def punctuate_text(input_text, progress=gr.Progress()):
+    """
+    Generate 3 different punctuation corrections with varying strategies
+    """
+    if not input_text.strip():
+        return ["", "", ""]
+
+    corrections = []
+
+    # Three different approaches
+    configs = [
+        {"name": "Conservative", "apply_sbd": False},
+        {"name": "With Sentence Boundaries", "apply_sbd": True},
+        {"name": "Balanced", "apply_sbd": True}
+    ]
+
+    for i, config in enumerate(configs):
+        progress((i + 0.5) / 3, desc=f"Generating {config['name']} version...")
+
+        if config["name"] == "Conservative":
+            # Single text processing without sentence boundaries
+            result = model.infer(texts=[input_text], apply_sbd=config["apply_sbd"])
+            corrected_text = result[0]
+
+        elif config["name"] == "With Sentence Boundaries":
+            # Process with sentence boundary detection
+            result = model.infer(texts=[input_text], apply_sbd=config["apply_sbd"])
+            corrected_text = "\n".join(result[0]) if isinstance(result[0], list) else result[0]
+
+        else:  # Balanced
+            # Process text in chunks if it's long
+            if len(input_text) > 500:
+                # Split into chunks
+                chunks = [input_text[i:i+500] for i in range(0, len(input_text), 400)]
+                results = []
+                for chunk in chunks:
+                    chunk_result = model.infer(texts=[chunk], apply_sbd=False)
+                    results.append(chunk_result[0])
+                corrected_text = " ".join(results)
+            else:
+                result = model.infer(texts=[input_text], apply_sbd=config["apply_sbd"])
+                corrected_text = "\n".join(result[0]) if isinstance(result[0], list) else result[0]
+
+        corrections.append(corrected_text)
+        progress((i + 1) / 3, desc=f"{config['name']} version complete")
+
+    progress(1.0, desc="All corrections generated!")
+    return corrections
+
+# Create Gradio interface
+with gr.Blocks(title="Multilingual Punctuation & Capitalization Correction", theme=gr.themes.Soft()) as demo:
+    gr.Markdown("""
+    # 🌍 Multilingual Punctuation & Capitalization Correction
+
+    This tool uses **XLM-RoBERTa** to restore punctuation, fix capitalization, and detect sentence boundaries in **47 languages**.
+
+    Enter text without proper punctuation or capitalization, and get 3 different correction styles:
+    - **📝 Conservative**: Minimal changes, preserves original flow
+    - **📖 With Sentence Boundaries**: Splits text into clear sentences
+    - **⚖️ Balanced**: Smart chunking for longer texts
+    """)
+
+    with gr.Row():
+        with gr.Column(scale=2):
+            input_text = gr.Textbox(
+                label="Input Text (any of 47 supported languages)",
+                placeholder="enter text without punctuation or capitalization like this example here it will be fixed",
+                lines=12,
+                max_lines=20
+            )
+            correct_btn = gr.Button("🚀 Add Punctuation & Capitalization", variant="primary", size="lg")
+
+    # Output section with 3 versions
+    gr.Markdown("### 📝 Correction Options")
+
+    with gr.Row():
+        with gr.Column():
+            gr.Markdown("#### 📝 Conservative")
+            output_conservative = gr.Textbox(
+                label="Conservative Correction",
+                lines=10,
+                max_lines=15,
+                interactive=True,
+                elem_id="conservative_output"
+            )
+            copy_btn_1 = gr.Button("📋 Copy", variant="secondary", size="sm")
+
+        with gr.Column():
+            gr.Markdown("#### 📖 With Sentence Boundaries")
+            output_boundaries = gr.Textbox(
+                label="Sentence Boundary Detection",
+                lines=10,
+                max_lines=15,
+                interactive=True,
+                elem_id="boundaries_output"
+            )
+            copy_btn_2 = gr.Button("📋 Copy", variant="secondary", size="sm")
+
+        with gr.Column():
+            gr.Markdown("#### ⚖️ Balanced")
+            output_balanced = gr.Textbox(
+                label="Balanced Correction",
+                lines=10,
+                max_lines=15,
+                interactive=True,
+                elem_id="balanced_output"
+            )
+            copy_btn_3 = gr.Button("📋 Copy", variant="secondary", size="sm")
+
+    # Selected version display
+    with gr.Row():
+        gr.Markdown("### ✅ Selected Correction")
+        selected_text = gr.Textbox(
+            label="Your Selected Correction",
+            lines=5,
+            interactive=True,
+            placeholder="Click 'Use This' under any correction to select it"
+        )
+
+    # Add selection buttons
+    with gr.Row():
+        with gr.Column():
+            select_btn_1 = gr.Button("✅ Use This", variant="primary", size="sm")
+        with gr.Column():
+            select_btn_2 = gr.Button("✅ Use This", variant="primary", size="sm")
+        with gr.Column():
+            select_btn_3 = gr.Button("✅ Use This", variant="primary", size="sm")
+
+    # Add examples
+    gr.Examples(
+        examples=[
+            ["hello there how are you doing today i hope everything is going well"],
+            ["the quick brown fox jumps over the lazy dog this is a test sentence"],
+            ["machine learning is revolutionizing many industries from healthcare to finance"],
+            ["bonjour comment allez vous aujourdhui jespere que tout va bien"],
+            ["hola como estas espero que todo este bien contigo y tu familia"],
+        ],
+        inputs=input_text,
+        label="Example sentences (click to try)"
+    )
+
+    # Set up event handlers
+    outputs = [output_conservative, output_boundaries, output_balanced]
+    correct_btn.click(fn=punctuate_text, inputs=input_text, outputs=outputs)
+    input_text.submit(fn=punctuate_text, inputs=input_text, outputs=outputs)
+
+    # Selection handlers
+    select_btn_1.click(fn=lambda x: x, inputs=output_conservative, outputs=selected_text)
+    select_btn_2.click(fn=lambda x: x, inputs=output_boundaries, outputs=selected_text)
+    select_btn_3.click(fn=lambda x: x, inputs=output_balanced, outputs=selected_text)
+
+    # JavaScript for copy functionality
+    copy_btn_1.click(
+        None,
+        None,
+        None,
+        js="""
+        () => {
+            const outputText = document.querySelector('#conservative_output textarea').value;
+            navigator.clipboard.writeText(outputText);
+            alert('Conservative version copied to clipboard!');
+        }
+        """
+    )
+
+    copy_btn_2.click(
+        None,
+        None,
+        None,
+        js="""
+        () => {
+            const outputText = document.querySelector('#boundaries_output textarea').value;
+            navigator.clipboard.writeText(outputText);
+            alert('Sentence boundaries version copied to clipboard!');
+        }
+        """
+    )
+
+    copy_btn_3.click(
+        None,
+        None,
+        None,
+        js="""
+        () => {
+            const outputText = document.querySelector('#balanced_output textarea').value;
+            navigator.clipboard.writeText(outputText);
+            alert('Balanced version copied to clipboard!');
+        }
+        """
+    )
+
+    gr.Markdown("""
+    ---
+    **Model:** [1-800-BAD-CODE/xlm-roberta_punctuation_fullstop_truecase](https://huggingface.co/1-800-BAD-CODE/xlm-roberta_punctuation_fullstop_truecase)
+
+    **Supports 47 languages** including: English, French, Spanish, German, Italian, Portuguese, Russian, Turkish, Chinese, Japanese, Arabic, and many more!
+    """)
+
+# Launch the app
+if __name__ == "__main__":
+    demo.launch(
+        server_name="0.0.0.0",
+        server_port=7860,
+        share=False
+    )

requirements.txt

@@ -0,0 +1,5 @@
+gradio>=4.0.0
+punctuators>=1.0.0
+torch>=2.0.0
+onnx>=1.14.0
+onnxruntime>=1.15.0