UseVoiceOptions — @personaai/react
Generated field table for UseVoiceOptions — do not edit
Generated from
sdk/react/dist/typedoc.json— do not edit.
UseVoiceOptions
Source: types.ts:448
Interface
| Field | Type | Description |
|---|---|---|
agentId? | string | Falls back to PersonaProvider's defaultAgentId if omitted. — _ types.ts:450_ |
threadId? | string | Resume an existing conversation over voice instead of starting fresh. start() mints the session with this thread id (same value useChat's threadId uses), so voice turns are persisted back into that thread's history — a later text message on the same thread sees them. Omit (or leave undefined) to start a fresh conversation. — _ types.ts:458_ |
contextOverride? | string | Caller-supplied context appended to the Agent's system instruction for this call — same field/cap/contract as useChat's sendMessage(text, { contextOverride }). Unlike text chat, this is a one-time append at connect, not re-applied per turn: Gemini Live's system instruction is fixed for the whole call, so changing it means ending this call and starting a new start(). Capped at 4000 characters server-side (rejected with a 400, not truncated). — _ types.ts:468_ |
context? | Record<string, unknown> | Caller-supplied context — never shown to the model, only ever read live by an RCP tool's resolver. This is the at-connect seed only, passed to start()'s ticket-mint call. To refresh it for the rest of an already-open call (a voice call can run up to 15 minutes — long enough for the caller's own context to change), use the returned updateContext() instead, which sends it live over the open WebSocket with no reconnect and no interruption to the audio. — _ types.ts:478_ |