Update app.py

app.py

@@ -11,7 +11,7 @@ description = "Gradio Demo for fairseq S2S: speech-to-speech translation models.
 article = "<p style='text-align: center'><a href='https://arxiv.org/abs/2107.05604' target='_blank'>Direct speech-to-speech translation with discrete units</a> | <a href='https://github.com/facebookresearch/fairseq/tree/main/examples/speech_to_speech' target='_blank'>Github Repo</a></p>"
 
 examples = [
-  ["enhanced_direct_s2st_units_audios_es-en_set2_source_12478_cv.flac",None,"xm_transformer_s2ut_800m-es-en-st-asr-bt_h1_2022"],
+  [None,"enhanced_direct_s2st_units_audios_es-en_set2_source_12478_cv.flac","xm_transformer_s2ut_800m-es-en-st-asr-bt_h1_2022"],
 ]
 
 io1 = gr.Interface.load("huggingface/facebook/xm_transformer_s2ut_800m-es-en-st-asr-bt_h1_2022", api_key=st.secrets["api_key"])
@@ -29,7 +29,7 @@ def inference(mic,file,model):
 
 gr.Interface(
     inference,
-    [gr.inputs.Audio(source="microphone", type="filepath", optional=True, label="Input"),gr.inputs.Audio(source="upload", type="filepath", optional=True, label="Input"),gr.inputs.Dropdown(choices=["xm_transformer_s2ut_800m-es-en-st-asr-bt_h1_2022"], type="value", label="model")
+    [gr.inputs.Audio(source="microphone", type="filepath", optional=True, label="Input"),gr.inputs.Audio(source="upload", type="filepath", optional=True, label="Input"),gr.inputs.Dropdown(choices=["xm_transformer_s2ut_800m-es-en-st-asr-bt_h1_2022"], default="xm_transformer_s2ut_800m-es-en-st-asr-bt_h1_2022",type="value", label="model")
 ],
     gr.outputs.Audio(label="Output"),
     article=article,