app.py

import gradio as gr
from whisper_model import transcribe_audio
from model import generate_prompt, generate_image_from_prompt

def gradio_interface_with_audio(audio_file):
    """ Gradio 界面函数，支持语音输入 """
    description = transcribe_audio(audio_file)
    # 生成提示
    prompt = generate_prompt(description)
    # 生成图像
    image = generate_image_from_prompt(prompt)
    return prompt, image

# 创建 Gradio 界面
iface = gr.Interface(fn=gradio_interface_with_audio, inputs="audio", outputs=["text", "image"])
iface.launch()