File size: 2,625 Bytes
a4a0cba 96644c9 a4a0cba f1d429a a4a0cba f1d429a a4a0cba f1d429a a4a0cba f1d429a a4a0cba f1d429a a4a0cba f1d429a 75961c1 f1d429a 75961c1 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 |
import streamlit as st
from PIL import Image
from transformers import pipeline
from gtts import gTTS
import torch
st.set_page_config(page_title="Image to Audio Story", page_icon="🦜")
# Load models once
caption_pipeline = pipeline("image-to-text", model="Salesforce/blip-image-captioning-base")
story_pipeline = pipeline("text-generation", model="Qwen/Qwen2-1.5B")
def extract_image_caption(image_data):
img_obj = Image.open(image_data)
caption_results = caption_pipeline(img_obj)
return caption_results[0]['generated_text']
def compose_story_from_caption(caption_detail):
prompt_text = (
"You are a talented and imaginative storyteller for children aged 3 to 10. "
"Using the details derived from the image below, craft a captivating tale that goes beyond merely describing the scene. "
"Let your creativity shine by introducing engaging characters, adventurous journeys, and delightful surprises. "
"Your story should be vivid, original, and between 100 and 300 words in length.\n\n"
f"Image Details: {caption_detail}\n\nStory:"
)
story_results = story_pipeline(prompt_text, num_return_sequences=1)
story_text = story_results[0]['generated_text']
return story_text.split("Story:", 1)[1].strip() if "Story:" in story_text else story_text.strip()
def convert_text_to_audio(text_content, audio_path="output.mp3"):
tts_engine = gTTS(text=text_content, lang="en")
tts_engine.save(audio_path)
return audio_path
def run_app():
st.markdown("<h1 style='text-align: center;'>Your Image to Audio Story 🦜</h1>", unsafe_allow_html=True)
st.write("Upload an image below and we will generate an engaging story from the picture, then convert the story into an audio playback!")
uploaded_image = st.file_uploader("Select an Image", type=["png", "jpg", "jpeg"])
if uploaded_image is not None:
image_display = Image.open(uploaded_image)
st.image(image_display, caption="Uploaded Image", use_container_width=True)
with st.spinner("Generating caption for the image..."):
caption_text = extract_image_caption(uploaded_image)
st.write("**Generated Caption:**", caption_text)
with st.spinner("Composing story..."):
story_text = compose_story_from_caption(caption_text)
st.write("**Story:**")
st.write(story_text)
with st.spinner("Converting text to audio..."):
audio_file = convert_text_to_audio(story_text)
st.audio(audio_file, format="audio/mp3")
if __name__ == "__main__":
run_app() |