Quick start overview

OtterVoice is a session orchestration SDK, not a model gateway. You pick a runtime, inject providers, and listen to session events.

Scenario → example

Scenario	Example	Command
Minimal loop (no mic / no key)	`examples/node-cli`	`bun run demo`
Node + live LLM	`examples/node-openrouter`	`bun run examples/node-openrouter/index.ts`
Browser full-duplex (native Audio LLM / server cascade)	`examples/web`	`cd examples/web && bun run start`
Browser Audio LLM + policy gateway	`examples/web-audio-llm-only`	`bun run examples/web-audio-llm-only/server.ts`
Expo app	`examples/react-native-expo`	`cd examples/react-native-expo && bun run start`
Scoped short-lived tokens (supported providers only)	`examples/token-broker`	`bun run examples/token-broker/server.ts`

Repo: github.com/bugkiwi/OtterVoice

Recommended path

Validate UI / captions / state with mocks.
Swap in the platform runtime (Web / Expo / Node).
Route request/response models through a server policy gateway. Use the token broker only when the provider can issue truly scoped credentials. Never ship long-lived keys, prompts, models, or spend controls to clients.

bun add @ottervoice/core
# Web:     bun add @ottervoice/runtime-web
# Expo:    bun add @ottervoice/runtime-react-native
# Node:    bun add @ottervoice/runtime-node
# Providers: @ottervoice/provider-deepgram | elevenlabs | openrouter | azure-speech