Text Summariser
Use this free online Text Summariser to pick the most important sentences from any article, report or document in seconds. It is useful for students preparing study notes, busy professionals digesting long emails, researchers scanning whitepapers, journalists pulling key points from press releases, and anyone with more reading than time. Paste your text into the box and pick a target length — short, medium or long — and the tool returns the highest-scoring sentences in their original order. Because the algorithm is extractive rather than generative, the wording is taken straight from the source, so quotation accuracy is preserved. Runs entirely in your browser, no sign-up required.
How It Works
This is an extractive summariser: it scores each sentence based on word frequency (TF) while ignoring common stopwords, then keeps the highest-scoring sentences in their original order. Best for factual articles. For abstractive AI summarisation, integrate an LLM via API.
How to Use This Text Summariser
Summarising a long article takes a few seconds.
- Paste the article or document into the input area.
- Pick the summary length — Short (~20%), Medium (~30%) or Long (~50%) of the original.
- Optionally adjust the minimum sentences slider so very short summaries always include at least N sentences.
- Click Summarise and wait for the highest-ranked sentences to appear.
- Click Copy to grab the summary for your notes.
How This Text Summariser Works
Automatic summarisation comes in two flavours. Abstractive summarisation uses a language model to generate new sentences that capture the meaning of the source — this is what services like ChatGPT do. Extractive summarisation, which is what this tool uses, picks the most important existing sentences instead of writing new ones. The algorithm works in three stages. First, the text is tokenised into sentences and words, with common stopwords like 'the', 'and', 'of' removed. Second, every remaining word receives a frequency score based on how often it appears in the document; rare words are valued less because they may be one-off names or numbers. Third, every sentence is scored as the average frequency of its remaining content words, and the top-scoring sentences are picked until the target length is reached. The selected sentences are then re-ordered to match the original document so the result reads coherently. Extractive summarisation works best on factual articles where keywords carry meaning. It is less effective on narrative writing, dialogue or text where the main idea is implicit — for those, an abstractive AI summariser will give better results.
Last updated: May 2026.
Frequently Asked Questions
How does extractive summarisation work?
Sentences are scored by the importance of the words they contain. The highest-scoring sentences are kept in their original order, producing a summary that uses the source's exact phrases.
Why are some important sentences missing?
Sentences with rare or unique vocabulary score low because their words appear only once in the text. Tweaking the summary length usually fixes this.
Will the summary be coherent?
Mostly — but because it skips sentences, transitions can feel abrupt. Read the result and adjust the length slider if it feels disjointed.
Can it summarise a PDF or web page?
Not directly. Copy the text content from the document and paste it into the input. For PDFs, use our PDF-to-Word tool to extract text first.
How does this differ from AI summarisation?
AI tools like GPT can rephrase and condense ideas in original wording. This tool simply selects existing sentences. The trade-off is privacy and speed — extractive is fully local.