Home › Text Tools › Text Summarizer

Text Summarizer

Paste an article, essay, report or email and get a short summary made of its most important sentences. Drag the slider to make the summary shorter or longer, and see which sentences were chosen and the main keywords. Your text never leaves your browser.

Text to summarize

0 words · 0 sentences

Summary length: 3 sentences

Length by

Format

Order

Top keywords

How to use the Text Summarizer

Paste your text into the box, or use the sample article to try it out. Longer texts (5+ sentences) give the best results.
Set the summary length with the slider. The summary updates instantly.
Pick the format: a paragraph or bullet points, in original order or ranked by importance.
Check the Highlighted tab to see which sentences were chosen in context, and the keyword list to see what the text is about.
Copy or download the summary as a .txt file.

How the summary is calculated (TF-IDF)

This summarizer uses extractive summarization: it never writes new sentences, it picks the most informative sentences from your text. Every step runs in your browser:

Split into sentences. The text is divided at ., ! and ?, while abbreviations such as Dr., e.g. and U.S., initials and decimal numbers are kept together. Short lines without punctuation are treated as headings and skipped.
Clean the words. Common function words (the, and, of, which…) are removed and words are reduced to a simple stem, so tree and trees count as the same term.
Weigh every term. Each term gets a TF-IDF weight: weight = TF × ln(1 + N ÷ DF), where TF is how often the term appears in the whole text, N is the number of sentences and DF is the number of sentences that contain the term. Frequent topic words score high; words that appear everywhere are dampened.
Score the sentences. A sentence's score is the sum of the weights of its distinct terms, divided by its length to the power 0.55 so long sentences don't win just by being long. The opening sentence gets a small bonus, because writers usually state the main idea first.
Pick without repeating. Sentences are taken from the highest score down. A sentence that shares most of its terms with one already chosen (cosine similarity above 0.62) is skipped, so the summary covers more ideas.

Tips for better summaries

Text type	Suggested length	Tip
News article	3 to 5 sentences	Original order reads most naturally.
Essay or blog post	15% to 25%	Bullet points make the key arguments easy to scan.
Research paper section	20% to 30%	Summarize one section at a time; leave out reference lists.
Long email or meeting notes	2 to 4 sentences	"Most important first" puts the main point on top.
Book chapter	5% to 10%	Check the keyword list to confirm the main themes.

Text copied from a PDF often has line breaks in the middle of sentences. The summarizer joins those lines automatically, but removing page numbers, headers and footnotes first gives cleaner results.

Frequently asked questions

How does this text summarizer work?

It is an extractive summarizer. The text is split into sentences, and every word gets a TF-IDF weight: words that appear often in the text but only in some sentences score high, while common words such as the or and are ignored. Each sentence is scored by the weight of its words, and the highest-scoring sentences that are not near-duplicates of each other form the summary.

What is the difference between extractive and abstractive summarization?

Extractive summarization copies the most important sentences word for word, so nothing is invented and every sentence can be checked against the original. Abstractive summarization, used by AI chatbots, writes new sentences; it can be more fluent but may introduce mistakes. This tool is extractive.

Is my text uploaded to a server?

No. The summary is calculated by JavaScript in your browser. Your text is not sent anywhere, so it is safe to use with private notes, school work or confidential documents.

How long should the summary be?

A common rule of thumb is 10% to 30% of the original length. For a news article, 3 to 5 sentences usually capture the main points. Use the slider to compare different lengths and check the Highlighted tab to see what is left out.

Which languages does it support?

Sentence splitting and scoring work for any language that uses spaces between words and normal sentence punctuation, such as Spanish, French, German or Portuguese. The list of ignored common words is English, so English texts give the best keywords.

Can I use the summary in my essay?

You can use it to understand or review a text, but the sentences are copied from the original author. If you include them in your own work, quote and cite the source, or rewrite the ideas in your own words.

Why is a sentence missing context in the summary?

Extractive summaries can pick a sentence that starts with a word like This or However, which refers to an earlier sentence. Increase the length slightly or switch to the Highlighted view to read it in context.

Related tools