duohub memory integration examples

This repository contains examples of how to use the duohub graph memory integration across different platforms and frameworks.

  Pipecat

This example shows how to use the duohub graph memory integration with the pipecat framework for voice AI.

It uses:

• OpenAI for the LLM
• Cartesia for the TTS
• Daily for the audio call
• duohub for the graph memory integration

Dependencies are managed with Poetry. The project can be containerised with Docker and deployed on ECS.

  Lambda

This example demonstrates how to integrate duohub's graph memory with AWS Lambda for serverless memory retrieval.

Available in both Python and TypeScript, it uses:

• AWS Lambda for serverless execution
• TypeScript/Python for type-safe development
• duohub for memory retrieval
• Axios for API requests (TypeScript)
• Requests for API calls (Python)

Dependencies are managed with npm/yarn (TypeScript) or pip (Python). The project can be deployed directly to AWS Lambda.

  Supabase Edge Functions

This example shows how to integrate duohub's graph memory with Supabase Edge Functions for serverless memory operations.

It uses:

• Supabase Edge Functions for serverless execution
• TypeScript/Deno runtime
• duohub for memory operations
• Built-in fetch for API requests

Dependencies are managed through Deno's import maps. The project can be deployed directly to Supabase using the CLI or GitHub Actions.

Project Structure