> ## Documentation Index
> Fetch the complete documentation index at: https://docs.deeda.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Voice & Realtime

> Speech in, speech out: what's driven from workflows vs the desktop app.

**ELI5:** voice has two homes. The **desktop app** owns live audio plumbing
(transports, turn detection, microphones). **Workflows** own *which agent
speaks with which voice* in multi-agent rooms. You declare the what; the app
runs the how.

## What the workflow author controls

In multi-agent rooms (`roster`), each agent can speak:

```yaml theme={null}
roster:
  - id: analyst
    runtime_key: openai-realtime-api
    voice: openai-realtime     # voice backend
    voice_name: cedar          # specific voice
  - id: narrator
    runtime_key: gemini-genai-sdk
    voice: gemini-tts
    voice_name: Kore
```

Single-agent workflows have **no voice field today** — a lone `agent` block
is text-first; voice sessions start from the desktop surfaces.

## What the app controls (not workflow-configurable)

Transport (WebSocket vs WebRTC), input/output modalities, and turn detection
(server VAD) are selected by the desktop runtime per session. SIP telephony
and custom vendor voices are **policy-blocked** (telephony trunk and vendor
program access deeda does not hold).

## What's implemented (probed)

OpenAI Realtime: WebSocket + WebRTC transports, in-session function tools,
remote MCP attachment, transcription-only sessions, input-audio
transcription, server-VAD turn detection, and the full client-event set.
Local voice: on-device STT with cloud-assisted TTS (Gemini TTS via gcloud
auth). See [OpenAI](/harness/providers/openai#realtime) and
[Local](/harness/providers/local) for row-level status.

## When to use what

| Task                                  | Route                                                                     |
| ------------------------------------- | ------------------------------------------------------------------------- |
| Voice conversation with one assistant | Desktop app voice surface (`local-voice` / realtime) — no workflow needed |
| Multi-expert spoken panel             | `roster` with per-agent `voice`/`voice_name`                              |
| Transcribe audio only                 | `openai-realtime-api` transcription session                               |
| Offline/private voice                 | `local-voice` (on-device STT; TTS needs gcloud auth)                      |

## See also

* [Runtimes](/harness/runtimes#openai-realtime-api) — knobs and starter
* [Workflow Schema](/harness/workflow-schema#roster) — roster field shapes