Summarize your audio data
In this guide, you'll learn how to use LeMUR to summarize your audio data with key takeaways.
If you want a Quickstart, see Apply LLMs to audio files.
To use LeMUR, you need an with a credit card set up.
With the LeMUR Task, you can send any prompt to the LLM and apply the model to your transcribed audio files.
Basic summarization example
To summarize the content in your audio data, define a summarization prompt and call transcript.lemur.task()
. The underlying transcript
is automatically used as additional context for the model.
Example output
The transcript describes several common sports injuries - runner's knee,
sprained ankle, meniscus tear, rotator cuff tear, and ACL tear. It provides
definitions, causes, and symptoms for each injury. The transcript seems to be
narrating sports footage and describing injuries as they occur to the athletes.
Overall, it provides an overview of these common sports injuries that can result
from overuse or sudden trauma during athletic activities
Custom summarization example
In this example, we'll use an advanced prompt to create custom summaries.
First, create a prompt with detailed summarization instructions:
You can also optionally specify a summary format and append it to the prompt.
Prompt the LeMUR model using transcript.lemur.task()
:
More summarization prompt examples
Try any of these prompts to get started:
Summaries Generate summaries for your audio data. | "Summarize key decisions and important points from the phone call transcript" |
Summarize audio segments Generate summaries for each segment or chapter of your audio data. | "Summarize the key events of each chapter" |
For more use cases and prompt examples, see LeMUR examples.
Improve the results
To improve the results, see the following resources:
- Optimize your prompt with the prompt engineering guide.
- To alter the outcome, see Customize LeMUR parameters.
- To get more deterministic and structured outputs for predefined tasks, see Specialized endpoints.