app.py

import streamlit as st
from transformers import pipeline






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)
    bytes_data = uploaded_file.getvalue()
    with open(uploaded_file.name, "wb") as file:
        file.write(bytes_data)
    st.image(uploaded_file, caption="Uploaded Image",
             use_column_width=True)

    #Define function:  
    def img2txt(imgname):
        pipe = pipeline("image-to-text", model="nlpconnect/vit-gpt2-image-captioning")
        scenario = pipe(imgname)
        return scenario[0]['generated_text']  

    def txt2story(txtname):
        pipe = pipeline("text-generation", model="openai-community/gpt2")
        story = pipe(txtname)
        return story[0]["generated_text"] 

    def text2audio(textname):
        pipe = pipeline("text-to-speech", model="facebook/mms-tts-eng")
        audio_data = pipe(textname)
        return audio_data  

    #Stage 1: Image to Text    
    st.text('Processing img2text...')
    scenario = img2txt(uploaded_file.name)
    st.write(scenario)

    #Stage 2: Text to Story
  
    st.text('Generating a story...')
    story = txt2story(scenario)
    st.write(story)

    #Stage 3: Story to Audio data  
    st.text('Generating audio data...')
    audio_data =text2audio(story)

    # Play button
    if st.button("Play Audio"):
        st.audio(audio_data['audio'],
                    format="audio/wav",
                    start_time=0,
                    sample_rate = audio_data['sampling_rate'])