In this guide, you’ll learn how to use AssemblyAI LeMUR to process an audio file and then automatically generate action items from the meeting’s transcript.

​
Overview​

Before we begin, make sure you have an AssemblyAI account and an API key. You can sign up for an AssemblyAI account and get your API key from your dashboard. LeMUR features are currently only available to paid users at two pricing tiers, LeMUR Default and LeMUR Basic. Refer to our pricing page for more detail.

​
Step-by-step instructions​

In this guide, you’ll use AssemblyAI to transcribe a meeting recording. Then, you’ll request a list of action items from LeMUR.
  1. Import the assemblyai package and set your API key
import assemblyai as aai
aai.settings.api_key = "YOUR_API_KEY"
  1. Use the Transcriber object’s transcribe method and pass in the audio file’s path as a parameter. The transcribe method will save the results of the transcription to the Transcribe object’s transcript attribute.
Python
transcriber = aai.Transcriber()
transcript = transcriber.transcribe("meeting.mp3")
  1. Define your LeMUR request context and answer_format parameters for the Action Items feature.
Python
# context is any information that would not be present in the transcript but may be helpful to know when generating action items
context = """This is a Product team meeting from 2019-07-09.
            Generate Action Items from the meeting in the following format:"""
# answer_format defines how you want the action items from LeMUR to be presented.
answer_format = """
            Action Item Title: A brief, descriptive title that summarizes the action item.
            Assignee: The person who is responsible for completing the action item.
            Due Date: The deadline for completing the action item.
            Status: The current status of the action item (e.g., "In progress", "Completed", "Deferred").
            Notes: Any additional notes or information about the action item."""
  1. Run the action_items method on transcript and print the result to your terminal.
Python
result = transcript.lemur.task(prompt)

print(result.response)