Update app.py

app.py

@@ -1,20 +1,25 @@
 import gradio as gr
 from prompt_generator import generate_prompt
 from image_generator import generate_image
+from whisper_model import transcribe_audio
 
-def generate_artwork(description, style, enhance_details):
+def generate_artwork(description, style, enhance_details, audio_file):
+    # 如果上传了音频文件，进行语音转文本
+    if audio_file is not None:
+        description = transcribe_audio(audio_file)
+    
     # 生成提示词
     prompt = generate_prompt(description)
     
     # 如果选中了增强细节的选项，修改提示
     if enhance_details:
-        prompt += " with enhanced details"
+        prompt += "有增强细节
     
     # 根据用户选择的风格修改提示
-    if style == "梦幻":
-        prompt += "以奇幻风格"
-    elif style == "Cyberpunk":
-        prompt += "赛博朋克"
+    if style == "奇幻":
+        prompt += " 奇幻风格"
+    elif style == "赛博朋克":
+        prompt += "赛博朋克风格"
     else:
         prompt += "写实风格"
 
@@ -25,17 +30,19 @@ def generate_artwork(description, style, enhance_details):
 # 创建 Gradio 界面
 with gr.Blocks() as demo:
     with gr.Row():
-        description_input = gr.Textbox(label="输入描述", placeholder="e.g., A magical treehouse in the sky")
-        style_selector = gr.Dropdown(choices=["梦幻", "赛博朋克", "写实"], label="选择风格")
+        description_input = gr.Textbox(label="请输入描述", placeholder="例如：天空中的魔法树屋")
+        style_selector = gr.Dropdown(choices=["奇幻", "赛博朋克", "现实主义"], label="选择风格")
         detail_checkbox = gr.Checkbox(label="增强细节")
+        audio_input = gr.Audio(source="microphone", label="录制您的描述", type="filepath")
         
     with gr.Row():
-        output_prompt = gr.Textbox(label="生成提示", interactive=False)
-        output_image = gr.Image(label="生成图像", interactive=False)
+        output_prompt = gr.Textbox(label="生成的提示词", interactive=False)
+        output_image = gr.Image(label="生成的图像", interactive=False)
 
     generate_button = gr.Button("生成作品")
     generate_button.click(generate_artwork, 
-                          inputs=[description_input, style_selector, detail_checkbox], 
+                          inputs=[description_input, style_selector, detail_checkbox, audio_input], 
                           outputs=[output_prompt, output_image])
 
 demo.launch()
+