Update app.py

app.py

@@ -134,7 +134,7 @@ def extract_speech_ids(speech_tokens_str):
             print(f"Unexpected token: {token_str}")
     return speech_ids
 
-
+@spaces.GPU(duration=15)
 def infer_with_speaker(
     display_name,
     target_text,
@@ -163,6 +163,9 @@ def infer_with_speaker(
 
 
 @spaces.GPU(duration=15)
+def gradio_infer(*args, **kwargs):
+    return infer(*args, **kwargs)
+
 def infer(
     sample_audio_path,
     target_text,
@@ -335,7 +338,7 @@ with gr.Blocks() as app_tts:
     )  # Add textbox
 
     generate_btn.click(
-        lambda *args: infer(*args, transcribed_text=None),
+        lambda *args: gradio_infer(*args, transcribed_text=None),
         inputs=[
             ref_audio_input,
             gen_text_input,