-
Notifications
You must be signed in to change notification settings - Fork 0
Home
peter-olai edited this page May 7, 2025
·
8 revisions
This wiki provides documentation for the Chat-Service, a backend service designed to integrate Large Language Models (LLMs), transcription, and Retrieval Augmented Generation (RAG) capabilities, for use with VR4VET NPCs and other AI functionalities.
See Setup-and-Installation to get started!
The Chat-Service provides the following core functionalities:
- Transcription: Converts spoken audio from users into text via Whisper.
- LLM Integration: Processes user transcribed audio and context to generate responses (via external provider).
- RAG Service: Enhances LLM responses by retrieving relevant information from a knowledge base. (mongodb Atlas)
- API Endpoints: Exposes these services for client applications to consume. (Blue-Sector)
Navigate through the sidebar to find more detailed information about specific components, setup, and API usage.
Authors:
Peter Olai Johnsen
Tobias Fremming
Erik Le Blanc Pleym
About Chat-Service
Core Components
Development
Codebase Details
- Codebase Architecture
- Routes
- Command-py
- Config-py
- Context Upload-py
- LLM-py
- Main-py
- Pipeline-py
- Streaming WS-py
- Transcribe-py
Deployment and Operations