Update app.py

app.py

@@ -1,75 +1,49 @@
+import gradio as gr
+from transformers import pipeline
 import speech_recognition as sr
-import pyttsx3
-from flask import Flask, jsonify, request
-import pandas as pd
+from gtts import gTTS
+import os
 
-# Initialize Flask app
-app = Flask(__name__)
+# Set up Hugging Face conversational model
+conversational_pipeline = pipeline("conversational", model="microsoft/DialoGPT-medium")
 
-# Initialize TTS engine
-engine = pyttsx3.init()
-engine.setProperty('rate', 150)  # Speed of speech
-
-# Sample restaurant menu
-menu = pd.DataFrame({
-    'ID': [1, 2, 3, 4],
-    'Name': ['Pizza', 'Burger', 'Pasta', 'Salad'],
-    'Price': [8.99, 5.49, 7.29, 3.99]
-})
-
-# Function to speak text
-def speak(text):
-    engine.say(text)
-    engine.runAndWait()
-
-# Function to listen for a command
-def listen():
+def process_audio(audio_file):
+    # Convert the audio file to text using SpeechRecognition
     recognizer = sr.Recognizer()
-    with sr.Microphone() as source:
-        print("Listening for your order...")
-        audio = recognizer.listen(source)
-    try:
-        command = recognizer.recognize_google(audio)
-        print("You said: ", command)
-        return command.lower()
-    except sr.UnknownValueError:
-        speak("Sorry, I did not understand. Can you please repeat?")
-        return listen()
-    except sr.RequestError:
-        speak("Could not request results. Please check your internet connection.")
-        return None
-
-# Function to process the order
-def process_order(command):
-    if 'order' in command:
-        food_item = None
-        for item in menu['Name']:
-            if item.lower() in command:
-                food_item = item
-                break
-        if food_item:
-            price = menu[menu['Name'] == food_item]['Price'].values[0]
-            speak(f"You've ordered {food_item}. The price is ${price}.")
-            return jsonify({'order': food_item, 'price': price})
-        else:
-            speak("Sorry, I couldn't find that item on the menu. Please try again.")
-            return None
-    else:
-        speak("Please say 'order' followed by the food item.")
-        return None
-
-@app.route('/start_order', methods=['GET'])
-def start_order():
-    speak("Welcome to the restaurant. What would you like to order?")
-    command = listen()
-    if command:
-        response = process_order(command)
-        if response:
-            return response
-        else:
-            return jsonify({'message': 'No valid order detected. Please try again.'}), 400
-    else:
-        return jsonify({'message': 'Error in voice recognition.'}), 500
-
-if __name__ == '__main__':
-    app.run(debug=True)
+    with sr.AudioFile(audio_file.name) as source:
+        audio = recognizer.record(source)
+        try:
+            print("Recognizing...")
+            text = recognizer.recognize_google(audio)
+            print(f"You said: {text}")
+        except sr.UnknownValueError:
+            text = "Sorry, I couldn't understand that."
+        except sr.RequestError:
+            text = "Could not request results."
+
+    # Get the bot's response using Hugging Face's model
+    response = conversational_pipeline(text)
+    bot_response = response[0]['generated_text']
+    print(f"Bot: {bot_response}")
+
+    # Convert the bot's response to speech using gTTS
+    tts = gTTS(bot_response)
+    tts.save("response.mp3")
+
+    # Play the audio
+    os.system("mpg321 response.mp3")
+    
+    return bot_response, "response.mp3"  # Return the bot's text response and the audio file
+
+# Create Gradio interface
+iface = gr.Interface(
+    fn=process_audio, 
+    inputs=gr.inputs.Audio(source="microphone", type="file"),
+    outputs=[gr.outputs.Textbox(), gr.outputs.Audio(type="file")],
+    live=True,
+    title="Voice Bot",
+    description="Speak to the bot, and it will respond to you!"
+)
+
+# Launch the interface
+iface.launch()