Update app.py

app.py

@@ -60,8 +60,8 @@ def generate_audio(text, model_id, language):
     output = pipe_dict["original_pipe"](text)
     output =  gr.Audio(value = (output["sampling_rate"], output["audio"].squeeze()), type="numpy", autoplay=True, label=f"Finetuned model prediction {default_model_per_language[language]}", show_label=True,
                                visible=True)
-    out.append(output)
-    out.extend([gr.Audio(visible=False)])
+    
+#    out.extend([gr.Audio(visible=False)])
     
     # if num_speakers>1:
     #     for i in range(min(num_speakers, max_speakers - 1)):
@@ -78,7 +78,7 @@ def generate_audio(text, model_id, language):
     #                            visible=True)
     #     out.append(output)
     #     out.extend([gr.Audio(visible=False)]*(max_speakers-2))
-    return out
+    return output
 
 
 css = """
@@ -115,10 +115,7 @@ with gr.Blocks(css=css) as demo_blocks:
                     )
                 
         with gr.Column():
-            outputs = []
-            for i in range(max_speakers):
-                out_audio = gr.Audio(type="numpy", autoplay=False, label=f"Generated Audio - speaker {i}", show_label=True, visible=False)
-                outputs.append(out_audio)
+            output = gr.Audio(type="numpy", autoplay=False, label=f"Generated Audio - speaker {i}", show_label=True, visible=False)
 
     with gr.Accordion("Datasets and models details", open=False):
         gr.Markdown("""
@@ -140,7 +137,7 @@ with gr.Blocks(css=css) as demo_blocks:
                     model_id
                    )
     
-    btn.click(generate_audio, [inp_text, model_id, language], outputs)
+    btn.click(generate_audio, [inp_text, model_id, language], output)
     
 
 demo_blocks.queue().launch()