app.py

# import part
import streamlit as st
from transformers import pipeline

# function part
# img2text
def img2text(url):
    image_to_text_model = pipeline("image-to-text", model="Salesforce/blip-image-captioning-base")
    text = image_to_text_model(url)[0]["generated_text"]
    return text

# text2story
def text2story(text):
    pipe = pipeline("text-generation", model="pranavpsv/genre-story-generator-v2")
    story_text = pipe(text)[0]['generated_text']
    return story_text

# text2audio
def text2audio(story_text):
    pipe = pipeline("text-to-audio", model="Matthijs/mms-tts-eng")
    audio_data = pipe(story_text)
    return audio_data


def main():
    st.set_page_config(page_title="Magic Story Box", page_icon="🧚")
    # 新版标题区
    st.markdown("""
    <div class="header">
        <h1>🪄 Magic Picture Story Box 🎧</h1>
        <h4 style="color:#4ECDC4;">Upload any photo to Get a fairy tale!</h4>
    </div>
    """, unsafe_allow_html=True)
    uploaded_file = st.file_uploader("🌈 Choose your magic picture...", type=["jpg", "png"])

    if uploaded_file is not None:
        # 保存上传文件（原有逻辑）
        bytes_data = uploaded_file.getvalue()
        with open(uploaded_file.name, "wb") as file:
            file.write(bytes_data)
        st.image(uploaded_file, caption="Your Magic Picture ✨", use_container_width=True)

        # 初始化状态容器（修复点：移出if块）
        status_container = st.empty()  # 移动到此处
        progress_bar = st.progress(0)

        # Stage 1: Image to Text
        with status_container.status("🔮 **Step 1/3**: Decoding picture magic...", expanded=True) as status:  # 保持缩进
            progress_bar.progress(33)
            scenario = img2text(uploaded_file.name)
            status.update(label="✅ Picture decoded!", state="complete")
            st.write(f"**What I see:** {scenario}")

        #Stage 2: Text to Story
        with status_container.status("📚 **Step 2/3**: Writing your fairy tale...", expanded=True) as status:
            progress_bar.progress(66)
            story = text2story(scenario)
            status.update(label="✅ Story created!", state="complete")
            st.write(f"**Your Story:**\n{story}")

        #Stage 3: Story to Audio data
        with status_container.status("🎵 **Step 3/3**: Adding magic music...", expanded=True) as status:
            progress_bar.progress(100)
            audio_data = text2audio(story)
            status.update(label="✅ All ready!", state="complete")


            # 自动播放（移除按钮）
            st.audio(audio_data['audio'], 
                    format="audio/wav",
                    start_time=0,
                    sample_rate=audio_data['sampling_rate'],
                    autoplay=True)  # 新增自动播放参数


if __name__ == "__main__":
    main()