Update app.py

app.py

@@ -355,11 +355,17 @@ if __name__ == "__main__":
                    "また、元音声と参考音声の合計時間が30秒を超える場合、元音声は分割処理されます。")
 
     with gr.Blocks() as demo:
+        gallery = gr.Gallery(
+            label="ギャラリー",
+            value=[item["webp"] for item in gallery_data],
+            interactive=True,
+            elem_id="gallery"
+        ).change(handle_gallery_selection, inputs=gr.Gallery, outputs="参考音声")  # Bind the correct inputs and outputs
+
         inputs = [
             gr.Audio(type="filepath", label="元音声"),
             gr.Audio(type="filepath", label="参考音声"),
-            gr.Gallery(label="ギャラリー", value=[item["webp"] for item in gallery_data],
-                       interactive=True, elem_id="gallery").change(handle_gallery_selection, inputs="gallery", outputs="参考音声"),
+            gallery,  # Add the gallery here
             gr.Slider(minimum=1, maximum=200, value=10, step=1, label="拡散ステップ数", info="デフォルトは10、50～100が最適な品質"),
             gr.Slider(minimum=0.5, maximum=2.0, step=0.1, value=1.0, label="長さ調整", info="1.0未満で速度を上げ、1.0以上で速度を遅くします"),
             gr.Slider(minimum=0.0, maximum=1.0, step=0.1, value=0.7, label="推論CFG率", info="わずかな影響があります"),
@@ -379,4 +385,4 @@ if __name__ == "__main__":
                      outputs=outputs,
                      title="Seed Voice Conversion",
                      cache_examples=False,
-                     ).launch()
+                     ).launch()