Eve library

Psyche Video Conferencing

Other

Revision 1

On this page

Video conferencing integration library for the Psyche AI Virtual Assistant Platform.

Part of the Oshun Platform.

Overview#

This library provides unified access to video conferencing platforms, enabling Psyche AI assistants to participate in video meetings with full media support.

Supported Platforms#

  • Zoom - Full API integration with meeting creation, joining, media streaming
  • Google Meet - Browser-based and API integration
  • Microsoft Teams - Bot framework and Graph API integration
  • Webex - API integration with meeting controls

Features#

Gateway Module#

  • Platform-agnostic meeting interface
  • Session management with reconnection
  • Recording controls
  • Participant management
  • Meeting scheduling and analytics

Meeting Intelligence#

  • Real-time topic analysis and summarization
  • Sentiment tracking per participant
  • Key moment detection
  • Follow-up tracking and action items
  • Conversation flow analysis

Multi-Participant Support#

  • Active speaker detection
  • Audio source separation
  • Gallery view tracking
  • Participant thumbnail extraction
  • Quality monitoring per participant

Perception#

  • Face analysis and emotion tracking
  • Body language detection
  • Screen content understanding
  • Multi-participant behavior analysis

Optimization#

  • Large meeting scaling (100+ participants)
  • Tiered processing based on importance
  • Adaptive sampling for performance
  • Resource allocation management

Virtual Devices#

  • Virtual camera for avatar rendering
  • Virtual audio device for TTS output
  • Platform-specific device integration

Project Structure#

text
src/video_conferencing/
├── __init__.py           # Package exports
├── gateway/              # Platform-agnostic interface
│   ├── interface.py      # Gateway protocol
│   ├── session.py        # Session management
│   ├── recording.py      # Recording controls
│   ├── participants.py   # Participant management
│   ├── scheduling.py     # Meeting scheduling
│   └── analytics.py      # Meeting analytics
├── zoom/                 # Zoom integration
├── meet/                 # Google Meet integration
├── teams/                # Microsoft Teams integration
├── webex/                # Webex integration
├── webrtc/               # WebRTC implementation
├── meeting_intelligence/ # AI-powered meeting analysis
├── multiparticipant/     # Multi-participant handling
├── perception/           # Visual/audio perception
├── optimization/         # Performance optimization
├── virtual_devices/      # Virtual camera/audio
└── accessibility/        # Captioning and A11y

Development#

Using Nx#

bash
# Install dependencies
nx install psyche-video-conferencing

# Build
nx build psyche-video-conferencing

# Linting
nx lint psyche-video-conferencing
nx lint-fix psyche-video-conferencing

# Formatting
nx format psyche-video-conferencing
nx format-check psyche-video-conferencing

# Type checking
nx typecheck psyche-video-conferencing

# Testing
nx test psyche-video-conferencing
nx test-cov psyche-video-conferencing

Direct Commands#

bash
cd apps/psyche/video-conferencing

# Install dependencies
poetry install

# Run tests
poetry run pytest

# Lint
poetry run ruff check src tests

Configuration#

Environment Variables#

Variable Description
ZOOM_CLIENT_ID Zoom OAuth client ID
ZOOM_CLIENT_SECRET Zoom OAuth client secret
ZOOM_WEBHOOK_SECRET Zoom webhook verification secret
GOOGLE_CLIENT_ID Google OAuth client ID
GOOGLE_CLIENT_SECRET Google OAuth client secret
TEAMS_CLIENT_ID Teams app client ID
TEAMS_CLIENT_SECRET Teams app client secret
WEBEX_CLIENT_ID Webex OAuth client ID
WEBEX_CLIENT_SECRET Webex OAuth client secret

Usage Example#

python
from video_conferencing import gateway, zoom

# Create a meeting session
session = await zoom.ZoomAdapter.join_meeting(
    meeting_id="123456789",
    password="password",
    display_name="Psyche Assistant",
)

# Enable virtual camera for avatar output
session.enable_virtual_camera()

# Start processing participants
async for event in session.participant_events():
    if event.type == "speaking":
        # Process speech from participant
        pass

License#

Proprietary - Oshun Platform