Dyslexia-Friendly Summarizer
A specialized text summarization model fine-tuned to produce summaries that are easier to read for people with dyslexia.
Model Details
- Developed by: Suyog Ghimire(Team DysIsfine)
- Model type: Text Summarization (Transformer-based)
- Language: English
- License: MIT
- Finetuned from: facebook/bart-large-cnn
What This Model Does
This model takes long texts and creates shorter, simpler summaries that are easier to read. It uses clear sentences and common words to make the summary accessible for people with dyslexia.
Example
Input: "The Sun is the main source of energy for Earth. It provides light and heat that support life. Plants use sunlight to make food, which provides energy for animals. The Sun also drives weather patterns and the water cycle. Without it, Earth would be too cold to support life."
Output: "The Sun provides light and heat that support life on Earth. Plants use sunlight to make food, and this energy passes to animals. Solar energy also drives weather and the water cycle, making Earth suitable for living organisms."
How to Use
from transformers import pipeline
summarizer = pipeline("summarization", model="[your-model-name]")
text = "Your long text here..."
summary = summarizer(text, max_length=100, min_length=30, do_sample=False)
print(summary[0]['summary_text'])
Training Data
The model was fine-tuned on a custom dataset of text-summary pairs. The training data includes educational content, news articles, and informational passages with simplified summaries.
Best For
- Educational content and learning materials
- News articles and informational texts
- Making long documents easier to understand
- Accessibility in web apps and reading tools
Limitations
- Works best with English text
- May lose some details when simplifying
- Should be reviewed by humans for important information (medical, legal, safety)
- Works better with moderate-length texts
Model Card Contact
- Downloads last month
- -