Update app.py

app.py

@@ -177,14 +177,13 @@ def process_image(image_file_path):
     except Exception as e:
         return f"Error occurred during image processing: {e}"
 
-def process_speech(audio_input, source_language, target_language="English"):
 
+def process_speech(audio_input, source_language, target_language="English"):
     if audio_input is None:
         return "No audio input provided."
     try:
-        # Predict using the client
         result = seamless_client.predict(
-            audio_input,  # File path of the audio
+            audio_input,
             source_language,
             target_language,
             api_name="/s2tt"
@@ -194,7 +193,6 @@ def process_speech(audio_input, source_language, target_language="English"):
         return f"Error in speech processing: {str(e)}"
 
 def convert_text_to_speech(input_text, source_language, target_language):
-
     try:
         result = seamless_client.predict(
             input_text,  
@@ -209,7 +207,6 @@ def convert_text_to_speech(input_text, source_language, target_language):
     except Exception as e:
         return None, f"Error in text-to-speech conversion: {str(e)}"
 
-
 def query_vectara(text):
     user_message = text
     customer_id = os.getenv('CUSTOMER_ID')
@@ -370,7 +367,7 @@ def process_and_query(input_language=None, audio_input=None, image_input=None, t
         # Process audio input
         elif audio_input is not None:
             audio_file_path = save_audio(audio_input)
-            audio_text = process_speech(input_language, audio_file_path)        
+            audio_text = process_speech(audio_file_path, input_language, "English")        
             combined_text += "\n\n**Audio Input:**\n" + audio_text
 
         # Process text input