app.py

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

# 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"]
}

# Context to track ongoing conversation
context = {"last_action": None, "order": []}

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

# Speech Recognition Function
def recognize_speech(audio_file_path):
    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):
    user_input = recognize_speech(audio_file_path)
    if user_input.startswith("Sorry") or user_input.startswith("I didn't"):
        return user_input, None

    global context

    # Default response
    response = f"You said: {user_input}\n"

    # Process the user input
    if "menu" in user_input.lower():
        response += "Here is our menu:\n"
        for category, items in menu.items():
            response += f"{category}: {', '.join(items)}\n"
        context["last_action"] = "show_menu"

    elif "order" in user_input.lower():
        if context["last_action"] == "show_menu":
            response += "What would you like to order?"
        else:
            response += "Please let me know what you'd like to order. We have starters, main course, breads, desserts, and drinks."
        context["last_action"] = "place_order"

    elif any(item.lower() in user_input.lower() for item in sum(menu.values(), [])):
        # Add the ordered item to the context
        for category, items in menu.items():
            for item in items:
                if item.lower() in user_input.lower():
                    context["order"].append(item)
                    response += f"I've added {item} to your order. Would you like anything else?"
                    break

    elif "no" in user_input.lower() or "done" in user_input.lower():
        if context["order"]:
            response += f"Your final order is: {', '.join(context['order'])}. Shall I place this order for you?"
            context["last_action"] = "confirm_order"
        else:
            response += "It seems you haven't ordered anything yet."

    elif "yes" in user_input.lower() and context["last_action"] == "confirm_order":
        response += f"Thank you! Your order for {', '.join(context['order'])} has been placed. Would you like to add a dessert or drink?"
        context["last_action"] = "offer_suggestions"

    elif any(item.lower() in user_input.lower() for item in menu["Desserts"] + menu["Drinks"]):
        # Add desserts or drinks to the order
        for item in menu["Desserts"] + menu["Drinks"]:
            if item.lower() in user_input.lower():
                context["order"].append(item)
                response += f"I've added {item} to your order. Your updated order is: {', '.join(context['order'])}. Thank you for ordering with us!"
                context["order"] = []  # Reset the order
                context["last_action"] = None
                break

    else:
        response += "I'm not sure what you meant. Could you clarify?"

    # Generate audio response
    audio_response_path = speak_and_save(response)
    return response, audio_response_path

# Gradio Interface
iface = gr.Interface(
    fn=process_order,
    inputs=gr.Audio(type="filepath"),  # Accept audio input as a file path
    outputs=[gr.Textbox(label="Text Response"), gr.Audio(label="Audio Response")],  # Include audio output
    title="Restaurant Voice Assistant",
    description="Talk to our voice assistant to place your order or ask about the menu!"
)

if __name__ == "__main__":
    iface.launch()