app.py

from transformers import pipeline
from langchain import PromptTemplate, LLMChain, OpenAI
import requests
import os
from dotenv import load_dotenv
import streamlit as st

load_dotenv(find_dotenv())
HUGGINGPHASE_API_TOKEN = os.getenv("HUGGINGPHASE_API_TOKEN")

# Image2Text

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

# LLM

def generate_story(scenario):
    template = """
    you are a story teller
    you can generate a short story based on simple narrative, the story should be more than 20 words;
    CONTEXT:{scenario},
    STORY:
    """
    prompt = PromptTemplate(template=template, input_variable=["scenario"])
    story_llm = LLMChain(llm=OpenAI(model_name="gpt-3.5-turbo", temperature=1), prompt=prompt, verbose=True)
    story = story_llm.predict(scenario=scenario)
    print(story)
    return story

# Text to Speech

def text2speech(message):
    API_URL = "https://api-inference.huggingface.co/models/espnet/kan-bayashi_ljspeech_vits"
    headers = {"Authorization": f"Bearer {HUGGINGPHASE_API_TOKEN}"}
    payloads = {
        "inputs": message
    }
    
    response = requests.post(API_URL, headers=headers, json=payloads)

    with open('audio.flac', 'wb') as file:
        file.write(response.content)

def main():
    st.set_page_config(page_title="img 2 Audio story", page_icon='🤖')
    st.header("Turn img into an audio story")
    uploaded_file = st.file_uploader("Choose an image...", type="jpg")

    if uploaded_file is not None:
        print(uploaded_file)
        with open(uploaded_file.name, "wb") as file:
            file.write(uploaded_file.getvalue())
        st.image(uploaded_file, caption="Uploaded Image.", use_column_width=True)
        scenario = img2text(uploaded_file.name)
        story = generate_story(scenario)
        text2speech(story)

        with st.expander("Scenario"):
            st.write(scenario)
        with st.expander("Story"):
            st.write(story)
        
        st.audio("audio.flac")

if __name__ == '__main__':
    main()