import streamlit as st
from transformers import pipeline
import hashlib

# Function definitions
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

def text2story(text):
    text_generation_model = pipeline("text-generation", model="aspis/gpt2-genre-story-generation")
    story_text = text_generation_model(
        text,
        min_length=50,
        max_length=100,
        do_sample=True,
        early_stopping=True,
        top_p=0.6
    )[0]["generated_text"]
    return story_text

def text2audio(story_text):
    text2audio_model = pipeline("text-to-speech", model="Matthijs/mms-tts-eng")
    gen_audio = text2audio_model(story_text)
    return gen_audio

def main():
    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:
        # Get file bytes and compute a hash
        bytes_data = uploaded_file.getvalue()
        file_hash = hashlib.sha256(bytes_data).hexdigest()
        
        # Reset session state only if the file content has changed, it prevents the regeneration after clicking "play audio"
        if ("last_uploaded_hash" not in st.session_state) or (st.session_state.last_uploaded_hash != file_hash):
            st.session_state.scenario = None
            st.session_state.story = None
            st.session_state.audio_data = None
            st.session_state.last_uploaded_hash = file_hash
        
        # Save the uploaded file locally.
        with open(uploaded_file.name, "wb") as file:
            file.write(bytes_data)
        st.image(uploaded_file, caption="Uploaded Image", use_container_width=True)
        
        # Stage 1: Image to Text
        if st.session_state.scenario is None:
            st.text("Processing img2text...")
            st.session_state.scenario = img2text(uploaded_file.name)
        st.write(st.session_state.scenario)
        
        # Stage 2: Text to Story
        if st.session_state.story is None:
            st.text("Generating a story...")
            st.session_state.story = text2story(st.session_state.scenario)
        st.write(st.session_state.story)
        
        # Stage 3: Story to Audio data
        if st.session_state.audio_data is None:
            st.text("Generating audio data...")
            st.session_state.audio_data = text2audio(st.session_state.story)
        
        # Play Audio button – uses stored audio_data.
        if st.button("Play Audio"):
            st.audio(
                st.session_state.audio_data["audio"],
                format="audio/wav",
                start_time=0,
                sample_rate=st.session_state.audio_data["sampling_rate"]
            )

if __name__ == "__main__":
    main()