Architecture¶

LaSuite Meet is a multi-service application. Understanding its components helps you deploy it correctly and debug issues when they arise.

Components overview¶

graph TB
    Browser["Browser\nReact / TypeScript"]
    OIDC["OIDC Provider\nKeycloak · Authentik · Google…"]

    subgraph backend_layer["Application"]
        Backend["Django Backend"]
        Celery["Celery Worker"]
    end

    subgraph media_layer["Media"]
        LiveKit["LiveKit Server"]
        Egress["LiveKit Egress"]
        Summary["Summary Service\n(optional - transcription & AI)"]
    end

    subgraph storage_layer["Storage"]
        PostgreSQL["PostgreSQL"]
        Redis["Redis"]
        ObjectStore["S3"]
    end

    Browser -->|REST API| Backend
    Browser -->|WebRTC| LiveKit
    Browser -->|OIDC login| OIDC
    OIDC -->|tokens| Backend
    Backend --> PostgreSQL
    Backend --> Redis
    Backend -->|Egress API| LiveKit
    Backend --> ObjcectStore
    Celery --> Redis
    LiveKit --> Redis
    Egress -->|writes recording| ObjectStore
    ObjectStore -->|storage webhook| Backend
    Summary -->|downloads from| ObjectStore
    Summary -->|reports results| Backend

Services¶

React Frontend¶

• Technology: TypeScript, React, Vite.js, React Aria (Adobe)
• Role: The browser-based UI. Connects to the Django backend via REST for authentication, room management, and recording state. Connects directly to LiveKit using the LiveKit JavaScript SDK for all real-time media (video, audio, screen sharing, chat).
• Port: 3000 (dev) / 8080 (production container)

Django Backend¶

• Technology: Python 3.13+, Django 5.x, Django REST Framework, Celery
• Role: Central API server. Handles:
• User authentication (OIDC/OAuth2)
• Room creation and access control
• Issuing LiveKit JWT tokens to clients
• Recording lifecycle management (start/stop/webhook)
• S3 webhook processing from ObjectStore
• Port: 8000

LiveKit Server¶

• Technology: Go (open source, by LiveKit Inc.)
• Role: The WebRTC Selective Forwarding Unit (SFU). Receives media streams from participants and selectively forwards them. Handles all real-time signaling, simulcast, codec negotiation, and TURN/STUN.
• Ports: 7880 (HTTP/WebSocket), 7881 (TCP), 7882/UDP (RTP/RTCP)

LiveKit Egress¶

• Technology: Go (open source, by LiveKit Inc.)
• Role: Records rooms or individual tracks to files. Saves output to ObjectStore. Triggered by the Django backend via LiveKit's Egress API.
• Dependency: Requires Redis (shared with LiveKit server)

Summary Service¶

• Technology: Python, FastAPI, Celery
• Role: Optional AI service for transcription and meeting summarization. Uses Whisper (or compatible STT engines) for speech-to-text, and an LLM API for summarization. Runs two separate Celery queues: transcribe-queue and summarize-queue.
• Port: 8000 (internal)

Metadata Collector Agent¶

• Technology: Python (LiveKit Agents SDK)
• Role: Connects to LiveKit rooms and collects metadata events: Voice Activity Detection (VAD), participant connection/disconnection events, chat messages. Stores metadata in object storage for later processing by the summary service.

Backing services¶

Authentication flow¶

sequenceDiagram
    participant B as Browser
    participant F as Meet Frontend
    participant BE as Meet Backend
    participant KC as OIDC Provider
    participant LK as LiveKit

    B->>F: Visit meet.example.com
    F->>BE: Check session
    BE-->>F: 401 - not authenticated
    F->>B: Redirect → OIDC provider
    B->>KC: Authenticate
    KC-->>B: Redirect → /api/v1.0/callback/?code=…
    B->>BE: GET /api/v1.0/callback/?code=…
    BE->>KC: Exchange code for tokens
    KC-->>BE: ID token + access token
    BE->>BE: Create / update user record
    BE-->>B: Session cookie, redirect → Meet
    B->>F: Load Meet (authenticated)
    F->>BE: POST /api/v1.0/rooms/{id}/token/
    BE-->>F: LiveKit JWT
    F->>LK: Connect via WebRTC

Recording flow¶

sequenceDiagram
    participant U as Room owner
    participant BE as Backend
    participant LK as LiveKit Server
    participant EG as LiveKit Egress
    participant S3 as ObjectStore
    participant Mail as SMTP

    U->>BE: Start recording
    BE->>LK: Create room composite egress
    LK->>EG: Spawn recording worker
    EG-->>LK: Join room, capture A/V
    U->>BE: Stop recording
    BE->>LK: Stop egress
    EG->>S3: Upload recording file
    LK-->>BE: Webhook: egress ended
    S3-->>BE: Webhook: file uploaded (/storage-hook/)
    BE->>BE: Set status → saved
    BE->>Mail: Send download link to owner
    Mail-->>U: Email with download link

Transcription flow (beta)¶

1. Recording completes (or transcription is started independently)
2. Django backend triggers the Summary service with the audio/video file location
3. Summary service downloads the file from object storage
4. Celery worker on transcribe-queue runs Whisper STT
5. Celery worker on summarize-queue calls the configured LLM API
6. Results are stored and made available for download via the Django API