Update app.py

app.py

@@ -18,27 +18,26 @@ def main():
     image = st.file_uploader("Upload an image", type=["jpg", "png"])
     question = st.text_input("Enter your question")
 
-    if image and question:
+    # 添加按钮以控制是否执行视觉问题回答
+    run_vqa = st.button('Run Visual Question Answering')
+
+    if image and question and run_vqa:
         # 将上传的文件转换为 PIL 图片
         image = Image.open(io.BytesIO(image.getvalue()))
 
         # 对用户上传的图片和问题进行视觉问题回答
         vqa_result = vqa_pipeline({"image": image, "question": question})
-        if vqa_result is None:  # 确保返回结果不为空
-            answer = "No answer here!"
-        else:
+        if vqa_result:  # 确保返回结果不为空
             answer = vqa_result[0]['answer']  # 获取回答
-        
-        st.write(f"Answer: {answer}")  # 显示回答
-
-        # 将回答转换为音频
-        tts_result = tts_pipeline(answer)
-        audio_data = tts_result['audio']  # 获取音频数据
-
-        # 添加一个按钮，让用户选择是否播放音频
-        if st.button('Play Answer Audio'):
-            st.audio(audio_data, format="audio/ogg")
-
+            st.write(f"Answer: {answer}")  # 显示回答
+
+            # 添加按钮以控制是否将回答转换为音频并播放
+            run_tts = st.button('Convert Answer to Audio')
+            if run_tts:
+                # 将回答转换为音频
+                tts_result = tts_pipeline(answer)
+                audio_data = tts_result['audio']  # 获取音频数据
+                st.audio(audio_data, format="audio/ogg")
 
 
 if __name__ == "__main__":