README.md

metadata

title: 'Multipurpose NLP: Translation and Summarization'
emoji: 🧠
colorFrom: blue
colorTo: green
sdk: gradio
sdk_version: 4.44.0
app_file: app.py
pinned: false

Multipurpose NLP Web App (Gradio on Hugging Face Spaces)

An interactive Gradio app offering:

• Multi-language text translation (M2M100: 100+ languages)
• Abstractive summarization in English (DistilBART CNN)

Demo

Once deployed, your Space will be live at: https://huggingface.co/spaces//

Features

• Translation: Choose source and target languages from 60+ common languages (backed by M2M100’s 100+ support).
• Summarization: Abstractive summaries of long text (best for English news/articles).

Models

• Translation: facebook/m2m100_418M
• Summarization: sshleifer/distilbart-cnn-12-6

Quick Start (Hugging Face Spaces)

1. Log in at https://huggingface.co
2. Create a new Space:
   • Space SDK: Gradio
   • Template: Blank (Gradio)
   • Choose a name (e.g., nlp-translate-summarize)
   • Hardware: CPU Basic (works; translation may be slower for long inputs)
3. Upload these files:
   • app.py
   • requirements.txt
   • README.md
4. Wait for the build to finish. The Space will auto-start.

Local Development (Optional)

python -m venv .venv && source .venv/bin/activate  # or .venv\Scripts\activate on Windows
pip install -r requirements.txt
python app.py
# Visit http://127.0.0.1:7860

Notes & Limitations

• Translation: For very long inputs, the model truncates to fit context (about ~1k tokens). Consider shorter paragraphs for best quality.
• Summarization: The chosen model is optimized for English. Non-English texts may yield weaker results.
• CPU-only inference is supported but slower; upgrading hardware on Spaces can improve latency.

License

MIT (adjust as needed)