Dialogue Summarization application hosted using AWS and CICD deployment with docker and FASTAPI. Model card created in HuggingFace and a deployed on HuggingFace Spaces.

Automating text summarization using machine learning techniques. Utilizing the Samsum dataset, our model generates concise summaries while retaining key information. Improves content comprehension and efficiency for information retrieval, content curation, and productivity enhancement.

The Text Summarizer project aims to distill the key points from lengthy articles, paragraphs, or conversations. Utilizing FastAPI, the project leverages Pegasus, a Google model, which operates on the Samsum dataset hosted on Hugging Face.

Successfully developed a text summarization model using Seq2Seq with attention to condense multi-turn dialogues from the SAMSum dataset into coherent and informative summaries.

Finetuned BART Model on the SAMSUM dataset and achieved average ROUGE scores: {'rouge1': 0.4872948285226925, 'rouge2': 0.25028497535417144, 'rougeL': 0.41239743837680054, 'rougeLsum': 0.41239743837680054}