Update app.py

app.py

@@ -1,311 +1,198 @@
 import gradio as gr
-import time
-import requests
-import io
-from PIL import Image
-from huggingface_hub import InferenceClient, HfApi
+from huggingface_hub import InferenceClient
 from deep_translator import GoogleTranslator
 from indic_transliteration import sanscript
 from indic_transliteration.detect import detect as detect_script
 from indic_transliteration.sanscript import transliterate
 import langdetect
 import re
-import os
-
-# Get secrets from Hugging Face Space
-HF_TOKEN = os.environ.get('HF_TOKEN')
-if not HF_TOKEN:
-    raise ValueError("Please set the HF_TOKEN secret in your HuggingFace Space")
 
 # Initialize clients
-text_client = InferenceClient(
-    "HuggingFaceH4/zephyr-7b-beta",
-    token=HF_TOKEN
-)
-
-# Image generation setup
-API_URL = "https://api-inference.huggingface.co/models/SG161222/RealVisXL_V4.0"
-headers = {"Authorization": f"Bearer {HF_TOKEN}"}
-
-def detect_language(text):
-    """Detect the language of input text"""
-    try:
-        return langdetect.detect(text)
-    except:
-        return 'en'
-
-def detect_and_transliterate(text):
-    """Detect script and transliterate if needed"""
-    try:
-        script = detect_script(text)
-        if script:
-            # Transliterate to Latin script if the text is in another script
-            return transliterate(text, script, sanscript.IAST)
-    except:
-        pass
-    return text
+text_client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
+image_client = InferenceClient("SG161222/RealVisXL_V3.0")
 
-def translate_to_english(text):
-    """Translate text to English if not already in English"""
+def detect_language_script(text: str) -> tuple[str, str]:
+    """Detect language and script of the input text.
+    Returns (language_code, script_type)"""
     try:
-        lang = detect_language(text)
-        if lang != 'en':
-            translator = GoogleTranslator(source=lang, target='en')
-            return translator.translate(text)
+        # Use confidence threshold to avoid false detections
+        lang_detect = langdetect.detect_langs(text)
+        if lang_detect[0].prob > 0.8:
+            # Only accept high confidence detections
+            lang = lang_detect[0].lang
+        else:
+            lang = 'en'  # Default to English if unsure
+        
+        script = None
+        try:
+            script = detect_script(text)
+        except:
+            pass
+        return lang, script
     except:
-        pass
-    return text
+        return 'en', None
 
-def check_for_image_generation(message):
-    """Check if message contains image generation commands"""
-    image_triggers = [
-        r"generate (?:an? )?image",
-        r"create (?:an? )?image",
-        r"draw",
-        r"show me",
-        r"visualize",
-        r"make (?:an? )?picture"
+def is_romanized_indic(text: str) -> bool:
+    """Check if text appears to be romanized Indic language.
+    More strict pattern matching."""
+    # Common Bengali romanized patterns with word boundaries
+    bengali_patterns = [
+        r'\b(ami|tumi|apni)\b',  # Common pronouns
+        r'\b(ache|achen|thako|thaken)\b',  # Common verbs
+        r'\b(kemon|bhalo|kharap)\b',  # Common adjectives
+        r'\b(ki|kothay|keno)\b'  # Common question words
     ]
     
-    for trigger in image_triggers:
-        if re.search(trigger, message.lower()):
-            return True
-    return False
+    # Require multiple matches to confirm it's actually Bengali
+    text_lower = text.lower()
+    matches = sum(1 for pattern in bengali_patterns if re.search(pattern, text_lower))
+    return matches >= 2  # Require at least 2 matches to consider it Bengali
 
-def user_message(message, history):
-    """Handle user messages"""
-    if message is None or message == "":
-        return "", history
+def translate_text(text: str, target_lang='en') -> tuple[str, str, bool]:
+    """Translate text to target language, with more conservative translation logic."""
+    # Skip translation for very short inputs or basic greetings
+    if len(text.split()) <= 2 or text.lower() in ['hello', 'hi', 'hey']:
+        return text, 'en', False
     
-    # Convert history to list if it's None
-    history = history or []
+    original_lang, script = detect_language_script(text)
+    is_transliterated = False
     
-    # Add user message to history with proper format
-    history.append({
-        "role": "user",
-        "content": message
-    })
+    # Only process if confident it's non-English
+    if original_lang != 'en' and len(text.split()) > 2:
+        try:
+            translator = GoogleTranslator(source='auto', target=target_lang)
+            translated = translator.translate(text)
+            return translated, original_lang, is_transliterated
+        except Exception as e:
+            print(f"Translation error: {e}")
+            return text, 'en', False
     
-    return "", history
-
-def bot_response(history, system_msg, max_tokens, temperature, top_p):
-    """Generate bot response"""
-    if not history:
-        return history, None
+    # Check for romanized Indic text only if it's a longer input
+    if original_lang == 'en' and len(text.split()) > 2 and is_romanized_indic(text):
+        text = romanized_to_bengali(text)
+        return translate_text(text, target_lang)  # Recursive call with Bengali script
     
+    return text, 'en', False
+
+def check_custom_responses(message: str) -> str:
+    """Check for specific patterns and return custom responses."""
+    message_lower = message.lower()
+    custom_responses = {
+        "what is ur name?": "xylaria",
+        "what is your name?": "xylaria",
+        "what's your name?": "xylaria",
+        "whats your name": "xylaria",
+        "how many 'r' is in strawberry?": "3",
+        "who is your developer?": "sk md saad amin",
+        "how many r is in strawberry": "3",
+        "who is ur dev": "sk md saad amin",
+        "who is ur developer": "sk md saad amin",
+    }
+    for pattern, response in custom_responses.items():
+        if pattern in message_lower:
+            return response
+    return None
+
+def is_image_request(message: str) -> bool:
+    """Detect if the message is requesting image generation."""
+    image_triggers = [
+        "generate an image",
+        "create an image",
+        "draw",
+        "make a picture",
+        "generate a picture",
+        "create a picture",
+        "generate art",
+        "create art",
+        "make art",
+        "visualize",
+        "show me",
+    ]
+    message_lower = message.lower()
+    return any(trigger in message_lower for trigger in image_triggers)
+
+def generate_image(prompt: str) -> str:
+    """Generate an image using DALLE-4K model."""
     try:
-        # Get the last user message
-        last_user_message = history[-1]["content"]
-        
-        # Process the message for language and script
-        processed_message = detect_and_transliterate(last_user_message)
-        translated_message = translate_to_english(processed_message)
-        
-        # Check for image generation request
-        if check_for_image_generation(translated_message):
-            # Generate image based on the message
-            image = generate_image(translated_message)
-            # Add bot response with proper format
-            history.append({
-                "role": "assistant",
-                "content": "I've generated an image based on your request."
-            })
-            return history, image
-        
-        # Prepare chat messages for the model
-        messages = [
-            {"role": "system", "content": system_msg},
-        ]
-        
-        # Add conversation history
-        for msg in history:
-            messages.append(msg)
-        
-        # Get model response
-        response = text_client.text_generation(
-            prompt=str(messages),
-            max_new_tokens=max_tokens,
-            temperature=temperature,
-            top_p=top_p,
-            repetition_penalty=1.1,
+        response = image_client.text_to_image(
+            prompt,
+            parameters={
+                "negative_prompt": "blurry, bad quality, nsfw",
+                "num_inference_steps": 30,
+                "guidance_scale": 7.5
+            }
         )
-        
-        # Add bot response with proper format
-        history.append({
-            "role": "assistant",
-            "content": response
-        })
-        
+        # Save the image and return the path or base64 string
+        # Note: Implementation depends on how you want to handle the image output
+        return response
     except Exception as e:
-        # Handle errors gracefully
-        error_message = f"I apologize, but I encountered an error: {str(e)}"
-        history.append({
-            "role": "assistant",
-            "content": error_message
-        })
+        print(f"Image generation error: {e}")
+        return None
+def romanized_to_bengali(text: str) -> str:
+    """Convert romanized Bengali text to Bengali script."""
+    bengali_mappings = {
+        'ami': 'আমি',
+        'tumi': 'তুমি',
+        'apni': 'আপনি',
+        'kemon': 'কেমন',
+        'achen': 'আছেন',
+        'acchen': 'আছেন',
+        'bhalo': 'ভালো',
+        'achi': 'আছি',
+        'ki': 'কি',
+        'kothay': 'কোথায়',
+        'keno': 'কেন',
+    }
     
-    return history, None
+    text_lower = text.lower()
+    for roman, bengali in bengali_mappings.items():
+        text_lower = re.sub(r'\b' + roman + r'\b', bengali, text_lower)
     
+    if text_lower == text.lower():
+        try:
+            return transliterate(text, sanscript.ITRANS, sanscript.BENGALI)
+        except:
+            return text
+            
+    return text_lower
+
 def create_chat_interface():
-    # Custom CSS for better styling with Inter font and animations
+    # Custom CSS for better styling
     custom_css = """
-    @import url('https://fonts.googleapis.com/css2?family=Inter:wght@400;500;600;700&display=swap');
-    
-    * {
-        font-family: 'Inter', sans-serif !important;
-    }
-    
-    .container {
-        max-width: 850px !important;
-        margin: auto;
-    }
-    
-    .chat-window {
-        height: 600px !important;
-        overflow-y: auto;
-        border-radius: 15px !important;
-        box-shadow: 0 8px 16px rgba(0, 0, 0, 0.1) !important;
-        transition: all 0.3s ease !important;
-    }
-    
-    .chat-window:hover {
-        box-shadow: 0 12px 20px rgba(0, 0, 0, 0.15) !important;
-    }
-    
-    .chat-message {
-        padding: 1rem !important;
-        margin: 0.5rem !important;
-        border-radius: 12px !important;
-        transition: all 0.2s ease !important;
-        opacity: 0;
-        animation: messageSlide 0.3s ease forwards;
-    }
-    
-    @keyframes messageSlide {
-        from {
-            opacity: 0;
-            transform: translateY(10px);
-        }
-        to {
-            opacity: 1;
-            transform: translateY(0);
-        }
-    }
-    
-    .user-message {
-        background: linear-gradient(135deg, #6366f1 0%, #8b5cf6 100%) !important;
-        color: white !important;
-        margin-left: 2rem !important;
-    }
-    
-    .bot-message {
-        background: linear-gradient(135deg, #f3f4f6 0%, #e5e7eb 100%) !important;
-        margin-right: 2rem !important;
-    }
-    
-    /* Button Styles */
-    button.primary {
-        background: linear-gradient(135deg, #6366f1 0%, #8b5cf6 100%) !important;
-        border: none !important;
-        color: white !important;
-        padding: 0.75rem 1.5rem !important;
-        border-radius: 12px !important;
-        font-weight: 600 !important;
-        transition: all 0.3s ease !important;
-        transform: translateY(0);
-        box-shadow: 0 4px 6px rgba(99, 102, 241, 0.2) !important;
-    }
-    
-    button.primary:hover {
-        transform: translateY(-2px);
-        box-shadow: 0 8px 12px rgba(99, 102, 241, 0.3) !important;
-    }
-    
-    button.primary:active {
-        transform: translateY(0);
-    }
-    
-    button.secondary {
-        background: #f3f4f6 !important;
-        border: 2px solid #e5e7eb !important;
-        color: #4b5563 !important;
-        padding: 0.75rem 1.5rem !important;
-        border-radius: 12px !important;
-        font-weight: 600 !important;
-        transition: all 0.3s ease !important;
-    }
-    
-    button.secondary:hover {
-        background: #e5e7eb !important;
-        border-color: #d1d5db !important;
-    }
-    
-    /* Input Styles */
-    .input-container {
-        position: relative;
-        margin-bottom: 1rem;
-    }
-    
-    textarea {
-        border: 2px solid #e5e7eb !important;
-        border-radius: 12px !important;
-        padding: 1rem !important;
-        transition: all 0.3s ease !important;
-        font-size: 1rem !important;
-        line-height: 1.5 !important;
-        box-shadow: 0 2px 4px rgba(0, 0, 0, 0.05) !important;
-    }
-    
-    textarea:focus {
-        border-color: #6366f1 !important;
-        box-shadow: 0 4px 6px rgba(99, 102, 241, 0.1) !important;
-    }
-    
-    /* Settings Panel */
-    .settings-block {
-        background: white !important;
-        border-radius: 15px !important;
-        padding: 1.5rem !important;
-        margin-top: 1rem !important;
-        box-shadow: 0 4px 6px rgba(0, 0, 0, 0.05) !important;
-        transition: all 0.3s ease !important;
-    }
-    
-    .settings-block:hover {
-        box-shadow: 0 6px 8px rgba(0, 0, 0, 0.08) !important;
-    }
-    
-    /* Slider Styles */
-    .gr-slider {
-        height: 4px !important;
-        background: #e5e7eb !important;
-        border-radius: 2px !important;
-    }
-    
-    .gr-slider .handle {
-        width: 16px !important;
-        height: 16px !important;
-        border: 2px solid #6366f1 !important;
-        background: white !important;
-        border-radius: 50% !important;
-        box-shadow: 0 2px 4px rgba(0, 0, 0, 0.1) !important;
-        transition: all 0.2s ease !important;
-    }
-    
-    .gr-slider .handle:hover {
-        transform: scale(1.1);
-    }
-    
-    /* Loading Animation */
-    @keyframes pulse {
-        0% { opacity: 1; }
-        50% { opacity: 0.5; }
-        100% { opacity: 1; }
-    }
-    
-    .loading {
-        animation: pulse 1.5s ease-in-out infinite;
-    }
+    body {
+    font-family: 'Inter', sans-serif;
+}
+
+.chat-container {
+    padding-top: 0;
+    padding-bottom: 0;
+}
+
+.chat-messages {
+    scroll-behavior: smooth;
+}
+
+.input-container {
+    border-top: 1px solid #ccc;
+}
+
+.input-container textarea {
+    border-radius: 12px 0 0 12px;
+}
+
+.input-container button {
+    border-radius: 0 12px 12px 0;
+}
+
+.loading {
+    animation: pulse 1.5s ease-in-out infinite;
+}
+
+@keyframes pulse {
+    0% { opacity: 1; }
+    50% { opacity: 0.5; }
+    100% { opacity: 1; }
+}
     """
 
     # Create the interface with custom theme
@@ -313,14 +200,37 @@ def create_chat_interface():
         # Header
         with gr.Row():
             gr.HTML("""
-                <div style="text-align: center; margin-bottom: 2rem; padding: 2rem;">
-                    <h1 style="font-size: 3rem; font-weight: 700; color: #4f46e5; margin-bottom: 0.5rem;">
-                        ✨ Xylaria Chat
-                    </h1>
-                    <p style="color: #6b7280; font-size: 1.2rem; font-weight: 500;">
-                        Your Intelligent Multilingual Assistant
-                    </p>
+                <!DOCTYPE html>
+<html lang="en">
+<head>
+    <meta charset="UTF-8">
+    <meta name="viewport" content="width=device-width, initial-scale=1.0">
+    <title>Xylaria Chat</title>
+    <link rel="stylesheet" href="styles.css">
+    <link rel="stylesheet" href="https://cdn.jsdelivr.net/npm/[email protected]/dist/full.css">
+    <script src="https://cdn.jsdelivr.net/npm/[email protected]/dist/vue.js"></script>
+</head>
+<body>
+    <div id="app" class="mx-auto max-w-5xl">
+        <header class="bg-white rounded shadow-lg p-10 text-center mb-10">
+            <h1 class="text-6xl font-bold text-violet-600 mb-5">✨ Xylaria Chat</h1>
+            <p class="text-lg font-medium text-gray-600">Your Intelligent Multilingual Assistant</p>
+        </header>
+        <section class="bg-white rounded shadow-lg p-10 chat-container">
+            <div class="chat-messages overflow-y-auto max-h-screen">
+                <div v-for="(message, index) in messages" :key="index" class="my-4" :class="{ 'flex justify-end': message.type === 'user' }">
+                    <div class="rounded-lg py-4 px-6" :class="{ 'bg-violet-500 text-white': message.type === 'user', 'bg-gray-200': message.type === 'bot' }">{{ message.text }}</div>
                 </div>
+            </div>
+            <div class="input-container flex mt-6">
+                <textarea v-model="inputText" class="w-full p-4 rounded-lg border-2 border-gray-400 resize-y" rows="3" placeholder="Type a message..."></textarea>
+                <button @click="sendMessage" class="bg-violet-500 text-white p-3 rounded-lg ml-4 hover:bg-violet-600 transition duration-300">Send</button>
+            </div>
+        </section>
+    </div>
+    <script src="script.js"></script>
+</body>
+</html>
             """)
 
         # Main chat interface
@@ -330,44 +240,26 @@ def create_chat_interface():
                     height=500,
                     show_label=False,
                     container=True,
-                    elem_classes=["chat-window"],
-                    type='messages'
-                )
-                
-                image_output = gr.Image(
-                    type="pil",
-                    label="Generated Image",
-                    visible=False,
-                    elem_classes=["generated-image"]
+                    elem_classes=["chat-window"]
                 )
                 
+                # Input area with buttons
+                with gr.Row():
+                    txt = gr.Textbox(
+                        show_label=False,
+                        placeholder="Type your message here...",
+                        container=False
+                    )
+                    send_btn = gr.Button("Send", variant="primary")
+                    clear_btn = gr.Button("Clear")
+
+                # Additional features bar
                 with gr.Row():
-                    with gr.Column(scale=8):
-                        txt = gr.Textbox(
-                            show_label=False,
-                            placeholder="Type your message here...",
-                            container=False,
-                            elem_classes=["input-textbox"]
-                        )
-                    with gr.Column(scale=1):
-                        send_btn = gr.Button(
-                            "Send",
-                            variant="primary",
-                            elem_classes=["primary"]
-                        )
-                    with gr.Column(scale=1):
-                        clear_btn = gr.Button(
-                            "Clear",
-                            variant="secondary",
-                            elem_classes=["secondary"]
-                        )
+                    audio_input = gr.Audio(source="microphone", type="filepath", label="Voice Input")
+                    image_output = gr.Image(label="Generated Image", visible=False)
 
-        # Settings panel
-        with gr.Accordion(
-            "⚙️ Advanced Settings",
-            open=False,
-            elem_classes=["settings-accordion"]
-        ):
+        # Settings panel (collapsible)
+        with gr.Accordion("Advanced Settings", open=False):
             with gr.Row():
                 with gr.Column():
                     system_msg = gr.Textbox(
@@ -398,11 +290,63 @@ def create_chat_interface():
                         label="Top-p (nucleus sampling)"
                     )
 
-        # Rest of your existing functions (user_message, bot_response, etc.)
-        # ... (keep the same function implementations)
+        # Function to handle sending messages
+        def user_message(message, history):
+            if message:
+                return "", history + [[message, None]]
+            return "", history
 
-        # Update the event handlers to use the new classes
-        send_event = txt.submit(
+        def bot_response(history, system_msg, max_tokens, temperature, top_p):
+            if len(history) == 0:
+                return history
+            
+            # Get the last user message
+            message = history[-1][0]
+            
+            # Check for custom responses first
+            custom_response = check_custom_responses(message)
+            if custom_response:
+                history[-1][1] = custom_response
+                return history
+            
+            # Check for image generation request
+            if is_image_request(message):
+                try:
+                    image = generate_image(message)
+                    if image:
+                        history[-1][1] = "Here's your generated image!"
+                        # Handle image display logic
+                        return history
+                except Exception as e:
+                    history[-1][1] = f"Sorry, I couldn't generate the image: {str(e)}"
+                    return history
+            
+            # Handle regular text responses
+            try:
+                translated_msg, original_lang, was_transliterated = translate_text(message)
+                response = respond(
+                    translated_msg,
+                    history[:-1],
+                    system_msg,
+                    max_tokens,
+                    temperature,
+                    top_p
+                )
+                
+                # Stream the response
+                partial_response = ""
+                for chunk in response:
+                    partial_response += chunk
+                    history[-1][1] = partial_response
+                    yield history
+                    time.sleep(0.02)  # Add slight delay for smooth streaming
+                
+            except Exception as e:
+                history[-1][1] = f"An error occurred: {str(e)}"
+                yield history
+
+        # Event handlers
+        txt_msg = txt.submit(
             user_message,
             [txt, chatbot],
             [txt, chatbot],
@@ -410,7 +354,7 @@ def create_chat_interface():
         ).then(
             bot_response,
             [chatbot, system_msg, max_tokens, temperature, top_p],
-            [chatbot, image_output]
+            chatbot
         )
 
         send_btn.click(
@@ -421,21 +365,20 @@ def create_chat_interface():
         ).then(
             bot_response,
             [chatbot, system_msg, max_tokens, temperature, top_p],
-            [chatbot, image_output]
+            chatbot
         )
 
-        clear_btn.click(
-            lambda: (None, None),
-            None,
-            [chatbot, image_output],
-            queue=False
-        )
+        clear_btn.click(lambda: None, None, chatbot, queue=False)
+
+        # Handle voice input
+        def process_audio(audio_file):
+            # Add your audio transcription logic here
+            return "Audio input received! (Add your transcription logic)"
 
-        # Update image visibility
-        send_event.then(
-            lambda img: gr.update(visible=img is not None),
-            image_output,
-            image_output
+        audio_input.change(
+            process_audio,
+            inputs=[audio_input],
+            outputs=[txt]
         )
 
     return demo