Meeting Summary: How to Get Decisions, Questions, and Tasks from Text
The Meeting Summary tool transforms meeting text into a structured list of decisions, open questions, and next actions. It works only with text that you manually enter or paste: two fields - 'Source material' (up to 8000 characters) and 'Goal or desired result' (up to 1000 characters). After pressing the Generate button you get a single result in a separate panel, which you can copy or download as plain text TXT. In this article we will explore how to prepare source text, formulate goals, check output, and consider limitations. All examples are illustrative and not real protocols.
Open tool: Meeting summary ↗Purpose of the Tool and Its Role in Protocol Work
Meeting Summary analyzes the provided text and extracts three categories: decisions made, open questions, and tasks. This helps you quickly see what was decided, what remained unclear, and what actions are anticipated. The tool does not connect to video conferencing systems, does not process audio or video, and does not perform speech recognition. It works exclusively with the text you entered.
Typical use case - rough protocol preparation or summary after a working meeting. You can use it for brainstorming, client discussions, or internal meetings. It is important to understand: the tool does not replace human judgment, it only structures information. Always check the output for consistency with the context.
The interface is extremely simple: two text fields, output language selection, and a Generate button. No presets, sliders, levels of changes, or additional settings exist. The result appears in a separate panel; it can be copied or downloaded as plain text TXT. History is not saved, automatic sending or publishing is not provided.
Source material: 'Anna will prepare a draft by Friday. Sam will check it on Monday. The launch date remains unresolved.' Goal: 'Highlight decisions, questions, and tasks'.
Preparing Source Text: Specificity Over General Phrases
For an accurate result, the source text must be coherent and contain explicit formulations. Avoid fragmentary phrases and pronouns without a specified subject. Instead of 'He will do it later', write 'Ivan will prepare a report later'. The more specific the formulations, the more reliably the tool will extract tasks and decisions. Indicate names or roles of responsible persons.
The text does not need to be perfectly formatted, but it is useful to separate different topics with paragraphs. If several issues were discussed, group related statements so the tool does not confuse the context. Remember the limit: the first field accepts up to 8000 characters. If the text is longer, shorten it, keeping the essence, or split it into parts and process separately.
In the 'Goal or desired result' field describe exactly what you want to get. For example: 'A list of tasks with responsible persons and deadlines' or 'Only open questions for the next meeting'. The field is optional, but it helps focus the output. If no goal is specified, the tool will give a general summary across the three categories.
Poor input: 'We discussed the project. We decided to improve it. Petya will do it.' Good input: 'We discussed project X. We decided to improve the main page design. Petya will prepare a layout by Wednesday'.
Step-by-Step Workflow with the Form
Step 1. Open the web form. In the 'Source material' field paste the meeting text, ensuring it does not exceed 8000 characters. If necessary, split the meeting into parts and process them separately. Step 2. In the 'Goal or desired result' field enter a specific goal, for example: 'Highlight decisions and tasks with responsible persons'. If the goal is not important, leave the field empty.
Step 3. Select the output language. Step 4. Click the Generate button. The tool will process the text and show the result in a separate panel. Step 5. Review the result: it will contain three sections - decisions, open questions, tasks. Tasks may include responsible persons and deadlines if they were in the text.
Step 6. Copy the result or download it as TXT. Note: the tool does not save history, does not send results by email, and does not integrate with other services. If the output is incomplete, edit the source text and repeat generation. The model may make errors, so always check the result.
After generation you may get: 'Decisions: main page design will be improved. Open questions: launch date not determined. Tasks: Petya will prepare a layout by Wednesday'.
Illustrative Example: Meeting on Workflow Improvement
Suppose you held a meeting with a team to discuss improving daily tasks. The text contained phrases such as: 'Anna will prepare a draft of the new regulation by Friday. Sam will check it on Monday. The implementation date remains unresolved. Also we need to decide who will be responsible for training employees'. You enter this text in the 'Source material' field and specify a goal: 'Highlight decisions, open questions, and tasks with responsible persons'.
The tool may give the following result: 'Decisions: regulation will be improved. Open questions: implementation date not determined; responsible person for training not assigned. Tasks: Anna will prepare a draft by Friday; Sam will check it on Monday'. Note: the tool did not add anything extra, it only structured the information from the text.
This example shows how the tool helps quickly create a protocol basis. However, it does not replace detailed analysis: if there were implicit agreements in the text, the tool may not recognize them. Therefore always compare with the source text and manually supplement the result if necessary.
Source text: 'Anna will prepare a draft by Friday. Sam will check in Monday. Launch date not resolved'. Possible result: 'Tasks: Anna - draft by Friday; Sam - check in Monday. Open question: launch date'.
Checking the Result: How to Distinguish Decision from Discussion
After generation carefully read each item. Make sure that decisions were actually made, not just discussed. Open questions should reflect topics on which no agreement was reached. Tasks should contain action verbs and, preferably, responsible persons and deadlines. If the tool missed something or distorted it, return to the source text and clarify formulations.
Check that questions that remained open did not end up in decisions. For example, the phrase 'We will think about this' is not a decision. Also make sure tasks do not duplicate or contradict each other. If the text contained numerical data or dates, verify them against the original. The model may make mistakes in details, especially if the text contains typos or ambiguous expressions.
Remember that the tool does not have contextual knowledge of your organization. It analyzes only what you entered. Therefore if there are references to previous meetings or external documents in the text, they may not be taken into account. In such cases add necessary clarifications to the source text or the goal field.
If the tool indicated 'Decided: move the meeting', but the text said 'Maybe we will move it?', this is an error. Correct manually.
Limitations and Warnings
The tool works only with text. It does not process audio, video, or images. If you have a meeting recording, first transcribe it to text manually or with another service. The tool does not perform OCR, does not recognize speech, and does not generate audio. It is purely a text analyzer.
Maximum source text length - 8000 characters, goals - 1000 characters. Exceeding the limit will cause an error. The tool does not save your data and has no history. Results are not sent by email and are not published automatically. You can only copy them or download as TXT. Other formats, such as DOCX or PDF, are not supported.
The model underlying the tool may make errors. It does not guarantee completeness or accuracy. Queue and availability may vary. Do not use the tool for making critical decisions without human verification. Also remember that the tool is not intended for legal, medical, or financial consultations. Always consult a specialist.
Do not try to upload an audio file - the tool will not accept it. First convert speech to text.
Practical Tips for Consistent Results
To get better results, follow simple rules. Write full names and roles. Use active constructions: 'Anna will do' instead of 'It will be done'. Explicitly indicate deadlines if they were discussed. Separate different topics with blank lines - this helps the tool correctly segment information.
If you are processing a long meeting, split it into logical parts and generate a result for each separately, then manually combine them. This avoids losing details due to character limits. You can also experiment with formulating the goal: the more specific the goal, the more precise the result. For example, 'Only tasks with responsible persons' will give a narrower output than 'All outcomes'.
Do not forget to check the result against the source text. The tool is an assistant, not a replacement for your attention. Use it for draft work, and compose the final protocol version yourself. This will save time and ensure accuracy.
For a long meeting: divide the text into blocks of 2000 characters, process each separately, then combine results into one document.
The Meeting Summary tool is a simple way to quickly structure meeting text. It extracts decisions, open questions, and tasks, helping you not miss important points. However, remember its limitations: text-only operation, character limits, possible model errors. Always check the result and use it as a basis for further work. By following our tips, you can effectively apply the tool in your daily practice.
Open tool: Meeting summary ↗Interface translations are machine-assisted and may need refinement.