Spaces:
Sleeping
Sleeping
| # Import necessary libraries | |
| import streamlit as st | |
| from transformers import pipeline | |
| from gtts import gTTS | |
| import os | |
| # Function to convert image to text using Hugging Face's BLIP model | |
| def img2text(url): | |
| """ | |
| Converts an image to text using the Salesforce/blip-image-captioning-base model. | |
| Args: | |
| url (str): Path to the image file. | |
| Returns: | |
| str: Generated text caption from the image. | |
| """ | |
| 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 | |
| # Function to generate a story from the text using T5-small model | |
| def text2story(text): | |
| """ | |
| Expands the text caption into a full story using the T5-small model. | |
| Args: | |
| text (str): Text caption generated from the image. | |
| Returns: | |
| str: Generated story suitable for kids aged 3-10. | |
| """ | |
| # Prompt to ensure the story is happy and suitable for kids | |
| prompt = f"Write a short, happy, and fun story for kids aged 3-10. The story should be about: {text}. Make it cheerful, imaginative, and suitable for young children. Avoid any scary or sad elements. The story should be under 95 words." | |
| text_generator = pipeline("text2text-generation", model="t5-small") | |
| story = text_generator(prompt, max_length=100)[0]['generated_text'] | |
| return story | |
| # Function to convert text to audio using gTTS | |
| def text2audio(story_text): | |
| """ | |
| Converts the generated story text into audio using gTTS. | |
| Args: | |
| story_text (str): The generated story text. | |
| Returns: | |
| str: Path to the generated audio file. | |
| """ | |
| tts = gTTS(text=story_text, lang='en') | |
| audio_file = "story_audio.mp3" | |
| tts.save(audio_file) | |
| return audio_file | |
| # Main application | |
| st.set_page_config(page_title="Your Image to Audio Story", | |
| page_icon="🦜") | |
| st.header("Turn Your Image into a Fun Audio Story!") | |
| uploaded_file = st.file_uploader("Select an Image...", type=["jpg", "jpeg", "png"]) | |
| if uploaded_file is not None: | |
| # Save the uploaded file | |
| bytes_data = uploaded_file.getvalue() | |
| with open(uploaded_file.name, "wb") as file: | |
| file.write(bytes_data) | |
| # Display the uploaded image | |
| st.image(uploaded_file, caption="Uploaded Image", use_column_width=True) | |
| # Stage 1: Image to Text | |
| st.text('Processing image to text...') | |
| scenario = img2text(uploaded_file.name) | |
| st.write("**Caption:**", scenario) | |
| # Stage 2: Text to Story | |
| st.text('Generating a fun story...') | |
| story = text2story(scenario) | |
| st.write("**Story:**", story) | |
| # Stage 3: Story to Audio | |
| st.text('Converting story to audio...') | |
| audio_file = text2audio(story) | |
| # Play button for the generated audio | |
| if st.button("Play Audio"): | |
| st.audio(audio_file, format="audio/mp3") | |
| # Clean up the generated audio file | |
| os.remove(audio_file) |