Update app.py

app.py

@@ -6,7 +6,6 @@ import gradio as gr
 # Global singleton: model loaded only once
 tts = None
 counter = 0  # To auto-number output files
-
 MODEL_NAME = "tts_models/multilingual/multi-dataset/xtts_v2"
 
 def load_model_once():
@@ -44,10 +43,12 @@ with gr.Blocks(title="XTTS Voice Clone (CPU)") as demo:
     out_audio = gr.Audio(label="Generated Voice")
     btn = gr.Button("Generate Voice")
 
+    # ⚡ Set concurrency_limit directly instead of deprecated queue()
     btn.click(
         fn=generate_audio,
         inputs=[text_in, wav_in],
-        outputs=out_audio
+        outputs=out_audio,
+        concurrency_limit=2  # adjust as needed
     )
 
-demo.queue(concurrency_count=1).launch()
+demo.launch(max_threads=2)  # adjust max_threads if needed