app.py

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

# Store cart in a temporary storage
cart = []

# Define the menu items dynamically
menu_items = {
    "Pizza": 10.99,
    "Burger": 8.49,
    "Pasta": 12.99,
    "Salad": 7.99,
    "Soda": 2.49
}

def generate_voice_response(text):
    tts = gTTS(text)
    temp_file = tempfile.NamedTemporaryFile(delete=False, suffix=".mp3")
    temp_file.close()
    tts.save(temp_file.name)
    return temp_file.name

def process_audio(audio_path):
    global cart
    recognizer = sr.Recognizer()
    response = ""

    try:
        with sr.AudioFile(audio_path) as source:
            audio = recognizer.record(source)
            input_text = recognizer.recognize_google(audio)
            print("User said:", input_text)

            if "menu" in input_text.lower():
                response = "Here is our menu:\n"
                for item, price in menu_items.items():
                    response += f"{item}: ${price:.2f}\n"
                response += "\nWhat would you like to add to your cart?"

            elif any(item.lower() in input_text.lower() for item in menu_items):
                for item in menu_items:
                    if item.lower() in input_text.lower():
                        cart.append(item)
                        total = sum(menu_items[cart_item] for cart_item in cart)
                        response = f"{item} has been added to your cart. Your current cart includes:\n"
                        for cart_item in cart:
                            response += f"- {cart_item}: ${menu_items[cart_item]:.2f}\n"
                        response += f"\nTotal: ${total:.2f}. Would you like to add anything else?"
                        break

            elif "final order" in input_text.lower() or "submit order" in input_text.lower():
                if cart:
                    total = sum(menu_items[cart_item] for cart_item in cart)
                    response = "Your final order includes:\n"
                    for item in cart:
                        response += f"- {item}: ${menu_items[item]:.2f}\n"
                    response += f"\nTotal: ${total:.2f}. Thank you for ordering!"
                    cart = []  # Clear cart after finalizing order
                else:
                    response = "Your cart is empty. Would you like to order something?"

            elif "stop" in input_text.lower():
                response = "Stopping voice assistant."

            else:
                response = "I didn’t quite catch that. Please tell me what you’d like to order."

    except sr.UnknownValueError:
        response = "Sorry, I didn’t catch that. Could you please repeat?"
    except Exception as e:
        response = f"An error occurred: {str(e)}"

    audio_response_path = generate_voice_response(response)
    return response, audio_response_path

def run_assistant(audio):
    transcription, audio_response_path = process_audio(audio)
    os.system(f"start {audio_response_path}")  # Automatically play the response audio
    return transcription

with gr.Blocks() as demo:
    gr.Markdown("# Voice-Activated Restaurant Assistant")
    audio_input = gr.Audio(type="filepath", label="Speak Now", streaming=True)

    output_text = gr.Textbox(label="Transcription")

    audio_input.change(fn=run_assistant, inputs=audio_input, outputs=output_text)

    demo.launch()