Update app.py

app.py

@@ -4,6 +4,9 @@ import os
 import tempfile
 import json
 import speech_recognition as sr
+import threading
+import time
+from playsound import playsound
 
 # Store cart in a temporary storage
 cart = []
@@ -17,6 +20,9 @@ menu_items = {
     "Soda": 2.49
 }
 
+# To manage playback control
+playback_control = {"stop": False, "pause": False}
+
 def generate_voice_response(text):
     tts = gTTS(text)
     temp_file = tempfile.NamedTemporaryFile(delete=False, suffix=".mp3")
@@ -27,73 +33,78 @@ def generate_voice_response(text):
 def calculate_total(cart):
     return sum(menu_items[item] for item in cart)
 
-def restaurant_voice_assistant(audio, state_json):
-    global cart
-    state = json.loads(state_json) if state_json else {}
-    response = ""
-    voice_path = None
-
-    # Convert audio input to text
-    if audio:
-        recognizer = sr.Recognizer()
-        with sr.AudioFile(audio) as source:
-            try:
-                input_text = recognizer.recognize_google(recognizer.record(source))
-            except sr.UnknownValueError:
-                input_text = ""
-    else:
-        input_text = ""
-
-    if not state.get("menu_shown", False):
-        # Show menu dynamically
-        response = "Welcome to our restaurant! Here is our menu:\n"
-        for item, price in menu_items.items():
-            response += f"{item}: ${price:.2f}\n"
-        response += "\nPlease tell me the item you would like to add to your cart."
-        state["menu_shown"] = True
-    elif any(item.lower() in input_text.lower() for item in menu_items):
-        # Check if input matches a menu item
-        for item in menu_items:
-            if item.lower() in input_text.lower():
-                cart.append(item)
-                total = calculate_total(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?"
+def play_audio_with_control(audio_path):
+    global playback_control
+    playback_control["stop"] = False
+    playback_control["pause"] = False
+    try:
+        for _ in range(3):  # Retry loop in case of interruptions
+            if playback_control["stop"]:
                 break
-    elif "menu" in input_text.lower():
-        response = "Here is our menu again:\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 "final order" in input_text.lower() or "submit order" in input_text.lower():
-        if cart:
-            total = calculate_total(cart)
-            response = "Your final order includes:\n"
-            for item in cart:
-                response += f"- {item}: ${menu_items[item]:.2f}\n"
-            response += f"\nTotal: ${total:.2f}.\nThank you for ordering!"
-            cart = []  # Clear cart after finalizing order
-        else:
-            response = "Your cart is empty. Would you like to order something?"
-    else:
-        response = "I didn’t quite catch that. Please tell me what you’d like to order."
+            playsound(audio_path)
+            break
+    except Exception as e:
+        print("Error playing audio:", e)
 
-    voice_path = generate_voice_response(response)
-    return response, voice_path, json.dumps(state)
+def restaurant_voice_assistant():
+    global cart
+    recognizer = sr.Recognizer()
 
-with gr.Blocks() as demo:
-    state = gr.State(value=json.dumps({}))
+    # Continuously listen for user commands
+    while True:
+        with sr.Microphone() as source:
+            print("Listening for input...")
+            try:
+                audio = recognizer.listen(source)
+                input_text = recognizer.recognize_google(audio)
+                print("You said:", input_text)
 
-    with gr.Row():
-        user_audio = gr.Audio(type="filepath", label="Your Voice Input")
-        output_text = gr.Textbox(label="Response Text")
+                # Process 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 = calculate_total(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 = calculate_total(cart)
+                        response = "Your final order includes:\n"
+                        for item in cart:
+                            response += f"- {item}: ${menu_items[item]:.2f}\n"
+                        response += f"\nTotal: ${total:.2f}.\nThank 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():
+                    playback_control["stop"] = True
+                    response = "Audio playback stopped."
+                elif "pause" in input_text.lower():
+                    playback_control["pause"] = True
+                    response = "Audio playback paused."
+                elif "resume" in input_text.lower():
+                    playback_control["pause"] = False
+                    response = "Resuming audio playback."
+                else:
+                    response = "I didn’t quite catch that. Please tell me what you’d like to order."
 
-    with gr.Row():
-        voice_output = gr.Audio(label="Response Audio", autoplay=True)
+                # Generate and play audio response
+                audio_path = generate_voice_response(response)
+                threading.Thread(target=play_audio_with_control, args=(audio_path,)).start()
 
-    submit_btn = gr.Button("Submit")
-    submit_btn.click(restaurant_voice_assistant, inputs=[user_audio, state], outputs=[output_text, voice_output, state])
+            except sr.UnknownValueError:
+                print("Sorry, I didn’t catch that. Could you please repeat?")
+            except Exception as e:
+                print("Error occurred:", e)
 
-demo.launch()
+# Run the assistant
+threading.Thread(target=restaurant_voice_assistant).start()