app.py

import gradio as gr
from gtts import gTTS
import os
import speech_recognition as sr
import time

# Menu for the restaurant
menu = {
    "Starters": ["Soup", "Spring Rolls"],
    "Main Course": ["Paneer Butter Masala", "Chicken Curry", "Veg Biryani"],
    "Breads": ["Roti", "Naan", "Paratha"],
    "Desserts": ["Gulab Jamun", "Ice Cream"],
    "Drinks": ["Mango Lassi", "Soda", "Water"]
}

conversation_history = []
context = {"last_action": None, "order": {}}

# Text-to-Speech Function
def speak_and_save(text):
    tts = gTTS(text=text, lang='en')
    audio_path = "response.mp3"
    tts.save(audio_path)
    return audio_path

# Speech Recognition Function
def recognize_speech(audio_file_path):
    if not audio_file_path or not os.path.exists(audio_file_path):
        return "No valid audio input provided. Please try again."
    
    recognizer = sr.Recognizer()
    try:
        with sr.AudioFile(audio_file_path) as source:
            audio_data = recognizer.record(source)
            text = recognizer.recognize_google(audio_data)
            return text
    except sr.UnknownValueError:
        return "I didn't catch that. Could you repeat?"
    except sr.RequestError:
        return "There was an issue processing your speech. Try again later."

# Voice Assistant Logic
def process_order(audio_file_path):
    if not audio_file_path or not os.path.exists(audio_file_path):
        response = "No audio input detected. Please record or upload your voice."
        return response, speak_and_save(response)

    user_input = recognize_speech(audio_file_path)
    conversation_history.append(f"Customer: {user_input}")
    
    if user_input.startswith("No audio input detected") or user_input.startswith("I didn't"):
        response = user_input
        conversation_history.append(f"AI: {response}")
        return response, speak_and_save(response)

    global context
    response = ""

    # Greet the customer
    if context["last_action"] is None:
        response = "Welcome to our restaurant. How can I assist you today?"
        context["last_action"] = "greet"
    elif "menu" in user_input.lower():
        response = "Here is our menu:\n"
        for category, items in menu.items():
            response += f"{category}: {', '.join(items)}\n"
        response += "What would you like to order?"
        context["last_action"] = "show_menu"
    elif "order" in user_input.lower() or any(item.lower() in user_input.lower() for item in sum(menu.values(), [])):
        words = user_input.split()
        quantity = 1  # Default quantity
        item_name = None
        for word in words:
            if word.isdigit():
                quantity = int(word)
            for category, items in menu.items():
                if word in items:
                    item_name = word
                    break
        
        if item_name:
            context["order"][item_name] = context["order"].get(item_name, 0) + quantity
            response = f"I've added {quantity} {item_name}(s) to your order. Would you like anything else?"
        else:
            response = "I couldn't identify the item. Could you please repeat your order?"
        context["last_action"] = "place_order"
    elif "no" in user_input.lower() or "that's it" in user_input.lower():
        order_summary = ", ".join([f"{item} (x{qty})" for item, qty in context["order"].items()])
        response = f"Your final order is: {order_summary}. Thank you for your order. Your food will arrive shortly. Enjoy your meal!"
        conversation_history.append(f"Order Summary: {context['order']}")
        context["last_action"] = "final_order"
        context["order"] = {}
    elif "everything alright" in user_input.lower() or "anything else" in user_input.lower():
        response = "I hope you're enjoying your meal. Can I bring you anything else?"
        context["last_action"] = "check_meal"
    else:
        response = "I'm not sure what you meant. Could you clarify?"
        context["last_action"] = "clarify"

    conversation_history.append(f"AI: {response}")
    audio_response_path = speak_and_save(response)
    return response, audio_response_path

# Gradio Interface with Auto-Play After 3 Seconds
def create_interface():
    with gr.Blocks() as app:
        gr.Markdown("## Restaurant Voice Assistant")
        gr.Markdown("Talk to our voice assistant to place your order or ask about the menu!")

        audio_input = gr.Audio(type="filepath", label="Your Voice Input")
        text_response = gr.Textbox(label="Text Response")
        audio_response = gr.Audio(label="Audio Response", interactive=False)

        # Add JavaScript for auto-playing audio 3 seconds after response
        gr.HTML("""
        <script>
        document.addEventListener("change", function() {
            setTimeout(function() {
                const audioPlayer = document.querySelector("audio");
                if (audioPlayer) {
                    audioPlayer.play();
                }
            }, 3000); // Delay of 3 seconds
        });
        </script>
        """)

        def interact(audio_file_path):
            return process_order(audio_file_path)

        audio_input.change(interact, inputs=audio_input, outputs=[text_response, audio_response])
    return app

if __name__ == "__main__":
    app = create_interface()
    app.launch()