Home icon

Meeting summarization and action item extraction with Amazon Nova

Machine Learning Blog



This article discusses using Amazon Nova models on Amazon Bedrock to automate meeting summarization and action item extraction using advanced prompt engineering techniques.

  • Key capabilities include generating high-level meeting summaries and structured action items
  • Amazon Nova offers four model tiers: Micro, Lite, Pro, and Premier with varying performance and efficiency
  • Evaluation used the QMSum dataset with 30 meeting transcripts
  • Performance testing revealed Nova Premier achieved highest faithfulness scores (1.0 for summarization, 0.83 for action items)
  • Smaller models like Nova Micro and Nova Lite offered faster processing times with slightly lower accuracy

The solution demonstrates how generative AI can efficiently transform unstructured meeting conversations into actionable insights, with Amazon Nova providing a scalable and cost-effective approach for enterprises.



Go to article

The AWS News Feed is currently looking for gold sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.

Related articles

Jul 10
2025
Building an AI-powered scientific meeting transcription platform with AWS
Jun 13
2025
Make videos accessible with automated audio descriptions using Amazon Nova
Jun 23
2025
Build an agentic multimodal AI assistant with Amazon Nova and Amazon Bedrock Data Automation
May 7
2024
Boost employee productivity with automated meeting summaries using Amazon Transcribe, Amazon SageMaker, and LLMs from Hugging Face

The AWS News Feed is currently looking for silver sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.