Enhance Your Public Speaking Skills with Amazon Bedrock Powered Virtual Assistant

Public speaking is a crucial skill in today’s world, whether for professional presentations, academic settings, or personal growth. By practicing regularly, individuals can build confidence, manage anxiety effectively, and develop strong communication skills for successful public speaking engagements. Now, with the emergence of large language models (LLMs), you can leverage generative AI-powered virtual assistants to receive real-time analysis of your speech, identify areas for improvement, and get suggestions for enhancing your delivery.

In this article, we introduce an Amazon Bedrock powered virtual assistant that can transcribe presentation audio and analyze it for language use, grammatical errors, filler words, and repetition. This virtual assistant provides recommendations and curated versions of the speech to elevate the presentation, helping refine communication skills and empowering individuals to become more effective public speakers. Organizations in various sectors can utilize this solution to provide automated coaching for their employees, students, and public speaking engagements.

Overview of the Solution

The solution comprises four main components:

• An Amazon Cognito user pool for authentication.
• A web portal created with Streamlit to upload audio and video recordings.
• A Step Functions workflow to convert audio to text using Amazon Transcribe and invoke Amazon Bedrock for speech recommendations and rewrite suggestions.
• Amazon Simple Notification Service (Amazon SNS) for sending email notifications with recommendations.

This solution utilizes Amazon Transcribe for speech-to-text conversion and Amazon Bedrock for generative AI capabilities to provide tailored recommendations. The integration of these technologies results in curated speech outputs that enhance presentation delivery.

For a detailed guide on constructing a scalable Public Speaking Mentor AI Assistant with Amazon Bedrock, Amazon Transcribe, and AWS Step Functions, refer to the sample code.

Prerequisites

Before implementing the solution, ensure you have:

1. An AWS account with appropriate IAM permissions for services like Amazon Bedrock, Amazon Transcribe, and others.
2. Access enabled for Anthropic’s Claude 3.5 Sonnet model on Amazon Bedrock.
3. A local desktop environment with Python, Git, and AWS CLI installed.

Deploying the Solution

Follow these steps to deploy the Public Speaking Mentor AI Assistant:

1. Clone the repository to your local desktop.
2. Navigate to the app directory in the cloned repository.
3. Create a Python virtual environment and install dependencies.
4. Deploy the CloudFormation template in your AWS account.

Using the Virtual Assistant

Once deployed, follow these steps to use the Public Speaking Mentor AI Assistant:

1. Open the provided Streamlit application URL in your browser.
2. Log in with your Amazon Cognito credentials.

Upload your audio or video recording, and the virtual assistant will process it to provide speech recommendations and rewrites. Access the results on the webpage and through email notifications via Amazon SNS.

Conclusion

Empower yourself as a captivating public speaker by utilizing the capabilities of LLMs in Amazon Bedrock. Enhance your communication skills with personalized feedback and constructive recommendations from the Public Speaking Mentor AI Assistant. Take the first step towards mastering the art of public speaking by trying out this cutting-edge technology today.

About the Authors

Nehal Sangoi and Akshay Singhal are Senior Technical Account Managers at Amazon Web Services, offering expert guidance on AWS solutions. Connect with them on LinkedIn and LinkedIn respectively.