app.py

from dotenv import find_dotenv, load_dotenv
from transformers import pipeline
import requests
import os
import streamlit as st

load_dotenv(find_dotenv())
api_token = os.getenv("HUGGINGFACEHUB_API_TOKEN")
#img2text
def img2text(url):
    image_to_text = pipeline("image-to-text",model='Salesforce/blip-image-captioning-large')

    text = image_to_text(url)[0]["generated_text"]
    #print(text)
    return text
#

#text2speech
def text2speech(message):
    API_URL = "https://api-inference.huggingface.co/models/espnet/kan-bayashi_ljspeech_vits"
    #API_URL = "https://api-inference.huggingface.co/models/microsoft/speecht5_tts"
    headers = {"Authorization": f"Bearer {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.title("Image to text to audio by 🤖")
    st.header("Turn image to audio podcast !!!")
    st.caption("Sample picture...")
    st.image("beachboat.jpg")
    img2text("beachboat.jpg")
    uploaded_file = st.file_uploader("Choose your image or simpley drag sample image given above",type="jpg")

    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)
        scenario = img2text(uploaded_file.name)
        text2speech(scenario)

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

if __name__ == '__main__':
    main()