Update app.py

app.py

@@ -1,56 +1,60 @@
 import gradio as gr
-from transformers import pipeline
 from indic_transliteration import sanscript
 from indic_transliteration.sanscript import transliterate
+from transformers import pipeline
 
-# Load Hugging Face emotion detection model
-emotion_pipeline = pipeline("text-classification", model="bhadresh-savani/bert-base-go-emotion")
-
-# Function: Transliterate + Emotion Detection
-def process_text(input_text):
-    # 1. Roman Hinglish → Devanagari Hindi
-    try:
-        hindi_text = transliterate(input_text, sanscript.ITRANS, sanscript.DEVANAGARI)
-    except Exception as e:
-        hindi_text = f"[⚠️ Transliteration Error: {str(e)}]"
-
-    # 2. Emotion Detection
-    try:
-        emotions = emotion_pipeline(input_text)  # original Hinglish gives better semantic results
-        # Convert to clean dictionary (label → score)
-        emotion_result = {emo['label']: round(emo['score'], 3) for emo in emotions}
-    except Exception as e:
-        emotion_result = {"Error": str(e)}
-
-    return hindi_text, emotion_result
-
-
-# ================== Gradio UI ==================
-with gr.Blocks(css=".gradio-container {background-color: #f9fafb;}") as demo:
-    gr.Markdown(
-        """
-        # 🔤 Hinglish → Hindi + 😃 Emotion Detection  
-        Enter your Hinglish text below and get:  
-        1. **Hindi Transliteration (Devanagari)**  
-        2. **Detected Emotions**  
-        ---
-        """
-    )
-
+# 1. Load Hindi Emotion Detection Model
+emotion_model = pipeline(
+    "text-classification",
+    model="s-nlp/HiEmotions",   # Hindi emotion detection model
+    return_all_scores=True
+)
+
+# 2. Basic Hinglish -> Hindi transliteration
+def hinglish_to_hindi(text: str) -> str:
+    return transliterate(text, sanscript.ITRANS, sanscript.DEVANAGARI)
+
+# 3. Normalization layer (common spelling fixes)
+def normalize_hindi(text: str) -> str:
+    corrections = {
+        "उदस्": "उदास",
+        "प्यर्": "प्यार",
+        "भाइयोन्": "भाइयों",
+        "कित्न": "कितना",
+        "टुझे": "तुझे"
+    }
+    for wrong, right in corrections.items():
+        text = text.replace(wrong, right)
+    return text
+
+# 4. Full pipeline
+def analyze_emotion(hinglish_text: str):
+    # Step 1: Transliterate Hinglish → Hindi
+    hindi_text = hinglish_to_hindi(hinglish_text)
+
+    # Step 2: Normalize Hindi text
+    normalized = normalize_hindi(hindi_text)
+
+    # Step 3: Predict Emotion
+    results = emotion_model(normalized)
+
+    # Prepare output
+    return {
+        "Input (Hinglish)": hinglish_text,
+        "Transliterated (Raw Hindi)": hindi_text,
+        "Normalized (Clean Hindi)": normalized,
+        "Predicted Emotions": results
+    }
+
+# 5. Gradio UI
+with gr.Blocks() as demo:
+    gr.Markdown("## Hinglish → Hindi Transliteration + Emotion Detection")
     with gr.Row():
-        input_text = gr.Textbox(
-            label="✍️ Enter Hinglish Text",
-            placeholder="e.g. Tujhe kitna pyar hai mujhse",
-            lines=2
-        )
-    
-    with gr.Row():
-        output_hindi = gr.Textbox(label="📝 Hindi Transliteration", interactive=False)
-        output_emotion = gr.JSON(label="🎭 Detected Emotions")
-
-    run_button = gr.Button("🚀 Run Analysis", variant="primary")
-    run_button.click(process_text, inputs=input_text, outputs=[output_hindi, output_emotion])
+        hinglish_in = gr.Textbox(label="Enter Hinglish Text", placeholder="e.g. Aam pataa nahi dil kuchh udas udas sa hai bhaaiyon")
+        output = gr.JSON(label="Results")
+    submit_btn = gr.Button("Analyze")
+    submit_btn.click(analyze_emotion, inputs=hinglish_in, outputs=output)
 
-# Run app
-demo.launch()
+if __name__ == "__main__":
+    demo.launch()