Update app.py

app.py

@@ -1,4 +1,6 @@
 import gradio as gr
+import torch
+import cv2
 import speech_recognition as sr
 from groq import Groq
 import os
@@ -6,9 +8,15 @@ import time
 import base64
 from io import BytesIO
 from gtts import gTTS
+import tempfile
 
 # Set device
+device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
+print(f"Using device: {device}")
 
+# Clear GPU memory if using GPU
+if torch.cuda.is_available():
+    torch.cuda.empty_cache()
 
 # Grok API client with API key (stored as environment variable for security)
 GROQ_API_KEY = os.getenv("GROQ_API_KEY", "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")  # Replace with your key or use env variable
@@ -113,10 +121,10 @@ def generate_response(user_input, emotion):
 def text_to_speech(text):
     try:
         tts = gTTS(text=text, lang='en', slow=False)
-        audio_buffer = BytesIO()
-        tts.write_to_fp(audio_buffer)
-        audio_buffer.seek(0)
-        return audio_buffer
+        # Create a temporary file to store the audio
+        with tempfile.NamedTemporaryFile(delete=False, suffix=".mp3") as temp_audio:
+            tts.save(temp_audio.name)
+            return temp_audio.name
     except Exception as e:
         print(f"Error generating speech: {str(e)}")
         return None
@@ -215,6 +223,6 @@ with gr.Blocks(theme=gr.themes.Soft(), css=css) as app:
         outputs=[chatbot, emotion_display, text_input, audio_output]
     )
 
-# Launch the app (for local testing; deployment will handle this differently)
+# Launch the app (for local testing)
 if __name__ == "__main__":
     app.launch(server_name="0.0.0.0", server_port=7860)