diff --git a/readme.txt b/readme.txt
new file mode 100644
index 0000000000000000000000000000000000000000..e13843982c03e181cec5419563e8647034f26224
--- /dev/null
+++ b/readme.txt
@@ -0,0 +1,29 @@
+Text Summarization Application
+
+The project focuses on developing a text summarization application that leverages state-of-the-art natural language processing (NLP) techniques. The key objectives are to efficiently extract text from various document formats, generate meaningful embeddings, and summarize large volumes of text data for improved readability and comprehension.
+Components and Workflow
+
+    Text Extraction:
+        The application extracts text from multiple sources, including PDF files and JSON documents. This involves reading the content of these files and converting it into a plain text format for further processing.
+
+    Text Embedding Generation:
+        Once the text is extracted, the application generates embeddings using a pre-trained language model. These embeddings transform the textual data into numerical representations that capture semantic meaning, facilitating similarity searches and further processing.
+
+    Similarity Search with FAISS:
+        To enable efficient retrieval of relevant documents based on content similarity, the project utilizes FAISS (Facebook AI Similarity Search) to create an index from the generated embeddings. This allows for fast querying and retrieval of similar documents based on user input.
+
+    Text Summarization:
+        The application integrates a summarization model from Hugging Face, specifically the facebook/bart-large-cnn model. This model is employed to generate concise summaries of the extracted texts, making it easier for users to grasp the essential points without reading through lengthy documents.
+
+    User Interaction:
+        Users can provide input queries or large blocks of text to be summarized. The application processes these inputs through the summarization chain, yielding succinct summaries that capture the core ideas presented in the original text.
+
+Use Cases
+
+    Information Retrieval: Quickly summarize and retrieve relevant information from extensive documents, such as research papers, reports, or legal agreements.
+    Content Consumption: Enable users to understand the main points of lengthy articles or papers without needing to read the entire content.
+    Data Analysis: Support analysts in distilling large datasets into manageable summaries for better decision-making and insights.
+
+Conclusion
+
+The project effectively combines text extraction, embedding generation, similarity search, and summarization into a cohesive application. By harnessing modern NLP techniques and tools, it addresses the challenges of information overload and enhances users' ability to digest and utilize textual information efficiently.
\ No newline at end of file