Refactor app.py to import 'spaces' and add GPU decorator; update requirements.txt to remove no-build-isolation flag from flash-attn

app.py

@@ -3,10 +3,8 @@ import torch
 from transformers import Qwen2_5OmniModel, Qwen2_5OmniProcessor
 from qwen_omni_utils import process_mm_info
 import soundfile as sf
-import os
-from datetime import datetime
 import tempfile
-import base64
+import spaces
 
 # Initialize the model and processor
 device = "cuda" if torch.cuda.is_available() else "cpu"
@@ -34,6 +32,7 @@ VOICE_OPTIONS = {
     "Ethan (Male)": "Ethan"
 }
 
+@spaces.GPU
 def process_input(user_input, chat_history, voice_type, enable_audio_output):
     # Prepare conversation history
     conversation = [SYSTEM_PROMPT]

requirements.txt

@@ -3,4 +3,4 @@ qwen-omni-utils[decord]
 soundfile
 torch
 gradio
-flash-attn --no-build-isolation
+flash-attn