Update app.py

app.py

@@ -93,8 +93,9 @@ def transcribe(audio_filepath):
 
 # Function to generate text using LLM
 def generate_text(input_text):
+    input_text = [input_text]
     generation_args = {
-        "max_new_tokens": 500,
+        "max_new_tokens": 300,
         "return_full_text": True,
         "temperature": 0.0,
         "do_sample": False,
@@ -119,8 +120,11 @@ def gen_speech(text):
 # Combined function for Gradio interface
 def process_audio(audio_filepath):
     transcription = transcribe(audio_filepath)
+    print("Done transcribing")
     generated_text = generate_text(transcription)
+    print("Done generating")
     audio_output_filepath = gen_speech(generated_text)
+    print("Done speaking")
     return transcription, generated_text, audio_output_filepath
 
 # Create Gradio interface