Workplace Conduct TrainingReal-time voice · STT → LLM → TTSworkplace.carecollaborative.cloud

Workplace Conduct Training Architecture

An AI persona joins a live WebRTC room and role-plays a workplace scenario — conduct review, performance feedback, conflict resolution. The trainee handles the conversation by voice in real time.

At a glance

What this use case is

Workplace Conduct Training places a trainee in a live voice conversation with an AI persona that role-plays difficult workplace situations. The trainee picks a scenario — a conduct review, a performance conversation, a conflict to resolve — and the AI agent joins the WebRTC room with scenario-specific character and emotional dynamics. The same real-time STT → LLM → TTS pipeline drives the persona's responses.

Interaction Model
Real-time voice simulation
Deployed At
workplace.carecollaborative.cloud
Data flow

Interaction pipeline

The ordered path a session takes through the platform for this use case.

1
Trainee voice → STT provider (real-time transcription)
2
STT output → AI Governance proxy → LLM (persona response)
3
LLM response → TTS provider (voice synthesis)
4
TTS audio → WebRTC room (returned to trainee)
Why it matters

Architecture highlights

Workplace scenarios with emotional dynamics
Provider-agnostic STT / LLM / TTS pipeline
Template-driven conduct assessment
Per-org Temporal namespace isolation
Service map

Platform capabilities

Internal Services
API Server (REST / DRPC / WS)
Temporal (durable orchestration)
Embedded WebRTC media server
AI Governance (MITM proxy)
Grading Engine (template-driven)
External Services (Provider-Agnostic)
Cloudflare (CDN / LB / Pages)
Identity Providers (SSO)
STT Providers (pluggable)
LLM Providers (pluggable)
TTS Providers (pluggable)
Go deeper

Explore the architecture

User Lifecycle — Workplace Conduct Training

Every phase, route, and protocol for this use case, end to end.