Update gradio_app_streaming.py

gradio_app_streaming.py

@@ -15,6 +15,7 @@ import threading
 from datetime import datetime
 from collections import deque
 from loguru import logger
+import spaces
 
 from flash_head.inference import (
     get_pipeline,
@@ -83,6 +84,7 @@ def _save_chunk_audio_to_wav(audio_array, wav_path, sample_rate=16000):
         wav_file.writeframes(samples.tobytes())
     return wav_path
 
+@spaces.GPU
 def run_inference_streaming(
     ckpt_dir,
     wav2vec_dir,