Update app.py

app.py

@@ -4,8 +4,8 @@ import pyttsx3
 import speech_recognition as sr
 
 # Replace with your Gemini API Key and endpoint
-API_KEY = "AIzaSyAm0RSpUKY38494Fug8SPIpdHLXE2d3cps"
-API_URL = "https://generativelanguage.googleapis.com/v1alpha2/models/gemini2:generateText"  # Example URL (adjust as needed)
+API_KEY = "AIzaSyAm0RSpUKY38494Fug8SPIpdHLXE2d3cps"  # Replace with your actual API key
+API_URL = "https://generativelanguage.googleapis.com/v1alpha2/models/gemini2:generateText"  # Adjust according to the Gemini API docs
 
 # Function to call Gemini API
 def call_gemini_api(message):
@@ -18,6 +18,7 @@ def call_gemini_api(message):
         "max_output_tokens": 100
     }
     try:
+        # Sending request to Gemini API
         response = requests.post(API_URL, headers=headers, json=payload)
         if response.status_code == 200:
             return response.json().get("generated_text", "No response text")
@@ -28,10 +29,15 @@ def call_gemini_api(message):
 
 # Convert text to speech (TTS)
 def text_to_speech(text):
-    engine = pyttsx3.init()
-    engine.save_to_file(text, "response.mp3")
-    engine.runAndWait()
-    return "response.mp3"
+    try:
+        engine = pyttsx3.init()
+        audio_filename = "response.mp3"
+        engine.save_to_file(text, audio_filename)
+        engine.runAndWait()
+        return audio_filename
+    except Exception as e:
+        print(f"Error with TTS: {e}")
+        return None
 
 # Convert audio to text (ASR)
 def audio_to_text(audio_path):
@@ -48,13 +54,17 @@ def audio_to_text(audio_path):
 # Define function for Gradio interface
 def respond(text_input=None, audio_input=None):
     if audio_input:
-        text_input = audio_to_text(audio_input)  # Convert audio to text if audio input is provided
+        # If audio input is provided, convert it to text
+        text_input = audio_to_text(audio_input)
     
     if not text_input:
         return "Error: No input provided.", None
     
-    api_response = call_gemini_api(text_input)  # Get response from Gemini API
-    audio_response = text_to_speech(api_response)  # Convert response text to audio
+    # Call Gemini API with text input and get response
+    api_response = call_gemini_api(text_input)
+    
+    # Convert the API response text into audio
+    audio_response = text_to_speech(api_response)
     
     return api_response, audio_response
 
@@ -72,4 +82,4 @@ demo = gr.Interface(
 )
 
 if __name__ == "__main__":
-    demo.launch(share=True)
+    demo.launch(debug=True)