Update app.py

app.py

@@ -42,24 +42,32 @@ def transcribe(audio_file):
 def generate_text(prompt):
     # formatted_prompt = format_prompt(prompt) # Removed
     # response = client.text_generation(formatted_prompt, max_new_tokens=250, temperature=0.7, top_p=0.95) # Removed
+    print(f"Generating text for prompt (type: {type(prompt)}): {prompt}") # Debug print
     result = llm_client.predict(query=prompt, api_name="/chat") # Added
+    print(f"Generated text result (type: {type(result)}): {result}") # Debug print
     return result.strip() # Modified to return the result from the Gradio Client
 
 # TTS Function
 def text_to_speech(text):
+    print(f"Converting text to speech (type: {type(text)}): {text}") # Debug print
     lang = detect(text)
     wav_path = "./output.wav"
     if lang == "sw":
         swahili_tts.synthesis(text, wav_path=wav_path)
     else:
         english_tts.synthesis(text, wav_path=wav_path)
+    print(f"TTS output path (type: {type(wav_path)}): {wav_path}") # Debug print
     return wav_path
 
 # Combined Processing Function
 def process_audio(audio):
+    print(f"Processing audio file (type: {type(audio)}): {audio}") # Debug print
     transcription = transcribe(audio)
+    print(f"Transcription result (type: {type(transcription)}): {transcription}") # Debug print
     generated_text = generate_text(transcription)
+    print(f"Generated text after function call (type: {type(generated_text)}): {generated_text}") # Debug print
     speech = text_to_speech(generated_text)
+    print(f"Speech output after function call (type: {type(speech)}): {speech}") # Debug print
     return transcription, generated_text, speech
 
 # Gradio Interface