Inspiration
Meetings are essential, but wasting time on long recordings and messy notes isn’t. That’s where SummarAIze comes in. We wanted to create a tool that automates meeting summarization, making it for people to stay aligned and productive.
What it does
SummarAIze is an AI-powered platform that seamlessly integrates with Google Meet, a video conferencing tool, and supports recordings. It automatically transcribes meetings, generates concise summaries, and extracts key insights, making it easier for users to review and understand their meetings.
How we built it
The frontend of SummarAIze was built using React to provide an interactive and responsive user interface. We integrated Three.js to create 3D visualizations and provide users with an engaging way to view their meetings. On the backend, we used FastAPI to handle user authentication, meeting management, and API calls to external services like Azure AI for transcription and summarization. We also set up a PostgreSQL database to store user data, meetings, and summaries. The Azure AI service was used for transcription via Azure Speech-to-Text, summarization using Azure Text Analytics, and storage using Azure Blob Storage.
Challenges we ran into
One of the key challenges was integrating with various video conferencing platforms like Google Meet, Zoom, and Microsoft Teams, as each platform has its own API and specific requirements for authentication, data retrieval, and access control. This created complexities in terms of managing different API endpoints, handling authorization mechanisms, and ensuring compatibility across all platforms. Due to these challenges, we ultimately decided to focus on integrating Google Meet initially, as it provided more accessible and comprehensive documentation, which allowed us to streamline the integration process. This decision helped us deliver a working solution faster while still providing valuable functionality.
Accomplishments that we're proud of
We successfully built a system that integrates with Google Meet, allowing users to upload meeting recordings, transcribe them, and generate summaries. The 3D visualizations using Three.js were a unique feature that provided an interactive way to explore meeting data. Additionally, the use of Azure AI to automate the transcription and summarization process was a significant accomplishment, enabling us to deliver valuable insights quickly and accurately.
What we learned
Throughout this project, we gained extensive knowledge of Azure AI services, particularly in the areas of transcription and text analytics. Azure's Speech-to-Text API allowed us to efficiently transcribe audio content from meeting recordings into text. This integration was crucial in automating the transcription process, enabling users to quickly access meeting content in text form.
What's next for SummarAIze
Moving forward, we plan to improve the platform by adding more integrations with additional platforms like Zoom and Microsoft Teams for better file sharing and collaboration. We also want to enhance the AI models to provide more accurate and insightful summaries. Additionally, we aim to expand the user interface to include more customizable features, such as the ability to set meeting preferences, filter meeting summaries, and provide deeper analytics on the meeting content.
Log in or sign up for Devpost to join the conversation.