Skip to main content

Summarizing virtual meetings

In this guide, we'll show you how to use the Summarization model to automatically generate summaries of your virtual meetings, so you can quickly review important information and take action based on the insights gathered from the conversations.

Get started

Before we begin, make sure you have an AssemblyAI account and an API key. You can sign up for a free account and get your API key from your dashboard.

The entire source code of this guide can be viewed here.

Step-by-step instructions

  1. 1

    Create a new file and import the necessary libraries for making an HTTP request.

  2. 2

    Set up the API endpoint and headers. The headers should include your API key.

  3. 3

    Upload your local file to the AssemblyAI API.

    We delete uploaded files from our servers either after the transcription has completed, or 24 hours after you uploaded the file. After the file has been deleted, the corresponding upload_url is no longer valid.

  4. 4

    Use the upload_url returned by the AssemblyAI API to create a JSON payload containing the audio_url parameter.

    To use the Summarization model, set summarization to True. By default, the summary_model model is informative and the summary_type is bullets.

    You can find all possible summary models and types here. Note that if the summary_model is specified, summary_type is also a required parameter, and vice versa.

  5. 5

    Make a POST request to the AssemblyAI API endpoint with the payload and headers.

  6. 6

    After making the request, you'll receive an ID for the transcription. Use it to poll the API every few seconds to check the status of the transcript job. Once the status is completed, you can retrieve the transcript from the API response, using the summary key to view the result of the summarization.

Understanding the response

Your transcript summary is located in the summary key of the API response. In addition to standard transcript metadata, you'll also find other parameters containing the summary_type, which indicates the type of summary that was generated, and summary_model, which indicates the specific AI summarization model that was used to generate the summary. More information on these parameters and their possible values can be found below.

Refer to the API reference for a breakdown of every element in your transcript output.

Best practices

Identify the most important information you want to include in the summary and choose the Summarization model and type that best suits your use case.

When using the conversational model, make sure that the two speakers have distinct voices and that there's minimal background noise. For shorter recordings, the gist or headline options may provide the most effective summary, whereas for longer recordings, the bullets_verbose or paragraph options may be more appropriate.

Consider the context in which the summary will be used, and whether additional context or explanation is necessary to understand the summary.

Experiment with different summarization options to find the one that works best for your specific use case.

Advanced usage

informative (default): Best for files with a single speaker such as presentations or lectures.bullets (default): A bulleted summary with the most important points.
conversational: Best for any two-person conversation such as customer/agent or interview/interviewee calls.bullets_verbose: A longer bullet point list summarizing the entire transcription text.
catchy: Best for creating video, podcast, or media titles.gist: A few words summarizing the entire transcription text.
headline: A single sentence summarizing the entire transcription text.
paragraph: A single paragraph summarizing the entire transcription text.
Learn more


AssemblyAI maintains a number of different summarization models for different use cases, so you can choose between a more informative summary or a catchier tagline for your audio. Visit the AssemblyAI blog for more examples of our summarization features in action.

If you're experiencing issues with generating a summary, make sure that you've included the summarization parameter in your request and set it to true. In addition, note that the Auto Chapters model and the Summarization model can't be active in the same request. If you try to enable both Auto Chapters and Summarization in a single request, you'll receive the following error message: "Only one of the following models can be enabled at a time: auto_chapters, summarization."