add print

app.py

@@ -28,11 +28,16 @@ def greet_json():
     return {"Hello": "World!"}
 
 # Function to generate audio from text using ParlerTTS
-def generate_audio(text, description="Neutral voice"):
+def generate_audio(text, description="Jon's voice is monotone yet slightly fast in delivery, with a very close recording that almost has no background noise."):
+    print("A")
     input_ids = tokenizer(description, return_tensors="pt").input_ids.to(device)
+    print("B")
     prompt_input_ids = tokenizer(text, return_tensors="pt").input_ids.to(device)
+    print("C")
     generation = model.generate(input_ids=input_ids, prompt_input_ids=prompt_input_ids)
+    print("D")
     audio_arr = generation.cpu().numpy().squeeze()
+    print("E")
     return audio_arr, model.config.sampling_rate
 
 # A POST endpoint to receive and parse an array of JSON objects
@@ -40,20 +45,23 @@ def generate_audio(text, description="Neutral voice"):
 async def create_items(items: List[Item]):
     processed_items = []
     for item in items:
+        print(f"Processing item: {item.text}")
         # Generate audio
+        print("before")
         audio_arr, sample_rate = generate_audio(item.text)
+        print("after")
 
-        # Create in-memory bytes buffer for audio
-        audio_bytes = io.BytesIO()
-        sf.write(audio_bytes, audio_arr, sample_rate, format="WAV")
-        audio_bytes.seek(0)  # Reset buffer position
+        # # Create in-memory bytes buffer for audio
+        # audio_bytes = io.BytesIO()
+        # sf.write(audio_bytes, audio_arr, sample_rate, format="WAV")
+        # audio_bytes.seek(0)  # Reset buffer position
 
         processed_item = {
             "text": item.text,
             "name": item.name,
             "section": item.section,
             "processed": True,
-            "audio": StreamingResponse(audio_bytes, media_type="audio/wav")
+            # "audio": StreamingResponse(audio_bytes, media_type="audio/wav")
         }
         processed_items.append(processed_item)