app.py

# import part
import streamlit as st
from transformers import pipeline
from gtts import gTTS
import os

# 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"]
        # Make the caption more fun and creative
        fun_caption = f"✨ Wow! Look at this! It’s a picture of {text.lower()}. Let’s turn it into a fun story! ✨"
        return fun_caption
    except Exception as e:
        st.error(f"Oops! Something went wrong while looking at your picture. Please try again! 🪄")
        return None

# text2story
def text2story(text):
    try:
        story_generator = pipeline("text-generation", model="gpt2")
        # Add a playful prompt to guide the story generation
        prompt = f"Once upon a time, there was {text}. "
        story = story_generator(prompt, max_length=95, num_return_sequences=1)[0]["generated_text"]
        # Make the story more fun by adding a happy ending
        fun_story = story + " And they all lived happily ever after! 🌈✨"
        return fun_story
    except Exception as e:
        st.error(f"Oops! Something went wrong while creating your story. Please try again! 🪄")
        return None

# text2audio
def text2audio(story_text):
    try:
        tts = gTTS(text=story_text, lang='en')
        audio_file = "story_audio.mp3"
        tts.save(audio_file)
        return audio_file
    except Exception as e:
        st.error(f"Oops! Something went wrong while turning your story into audio. Please try again! 🪄")
        return None

# main part
st.set_page_config(page_title="Story Maker", page_icon="🦄")
st.header("🌟 Story Maker: Turn Your Picture into a Fun Story! 🌟")
uploaded_file = st.file_uploader("📷 Choose a picture to create a fun story...", type=["jpg", "jpeg", "png"])

if uploaded_file is not None:
    bytes_data = uploaded_file.getvalue()
    with open(uploaded_file.name, "wb") as file:
        file.write(bytes_data)

    st.image(uploaded_file, caption="Your picture!", use_container_width=True)

    # Stage 1: Image to Text
    st.write("✨ Let’s see what’s in your picture... ✨")
    scenario = img2text(uploaded_file.name)
    if scenario:
        st.write(scenario)

        # Stage 2: Text to Story
        st.write("🧙‍♂️ Turning your picture into a story... 🧙‍♀️")
        story = text2story(scenario)
        if story:
            st.write("📖 Here’s your story:")
            st.write(story)

            # Stage 3: Story to Audio
            st.write("🎤 Getting ready to tell your story... 🎤")
            audio_file = text2audio(story)
            if audio_file:
                # Play button
                if st.button("🎧 Listen to Your Story!"):
                    st.audio(audio_file, format="audio/mp3")
                # Clean up the audio file after playing
                os.remove(audio_file)