import streamlit as st
import importlib.util
try:
    # 检查 accelerate 库是否安装
    spec = importlib.util.find_spec("accelerate")
    if spec is None:
        st.error("缺少 'accelerate' 库，请安装该库以加载 FP8 量化模型。可以使用 'pip install accelerate' 进行安装。")
        st.stop()
    from transformers import pipeline
    from gtts import gTTS
    import io
    import tempfile
    import os
except ImportError as e:
    st.error(f"导入库时出错: {e}")
    st.stop()

# function part
# img2text
def img2text(url):
    try:
        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
    except Exception as e:
        st.error(f"图像描述生成出错: {e}")
        return None

# text2story
def text2story(text):
    try:
        story_generator = pipeline("text-generation", model="deepseek-ai/DeepSeek-R1", trust_remote_code=True)
        story = story_generator(text, max_length=200, num_return_sequences=1)[0]['generated_text']
        return story
    except Exception as e:
        st.error(f"故事生成出错: {e}")
        return None

# text2audio
def text2audio(story_text):
    try:
        tts = gTTS(text=story_text, lang='en')
        audio_file = io.BytesIO()
        tts.write_to_fp(audio_file)
        audio_file.seek(0)
        return audio_file
    except Exception as e:
        st.error(f"文本转语音出错: {e}")
        return None

st.set_page_config(page_title="Your Image to Audio Story",
                   page_icon="🦜")
st.header("Turn Your Image to Audio Story")
uploaded_file = st.file_uploader("Select an Image...")

if uploaded_file is not None:
    print(uploaded_file)
    # 使用临时文件处理上传的图像
    with tempfile.NamedTemporaryFile(delete=False, suffix=".jpg") as temp_file:
        temp_file.write(uploaded_file.getvalue())
        temp_file_path = temp_file.name

    st.image(uploaded_file, caption="Uploaded Image",
             use_container_width=True)  # 修改为 use_container_width

    #Stage 1: Image to Text
    st.text('Processing img2text...')
    scenario = img2text(temp_file_path)
    if scenario:
        st.write(scenario)

        #Stage 2: Text to Story
        st.text('Generating a story...')
        story = text2story(scenario)
        if story:
            st.write(story)

            #Stage 3: Story to Audio data
            st.text('Generating audio data...')
            audio_data = text2audio(story)
            if audio_data:
                # Play button
                if st.button("Play Audio"):
                    st.audio(audio_data,
                             format="audio/mpeg",
                             start_time=0)

    # 删除临时文件
    os.remove(temp_file_path)