RimeRealtime
Speech-to-speech conversation with Rime Prism, using your own Rime API key. Choose it for a phone-ready model that finds turns and handles interruptions itself, with a short spoken phrase while tools run. Authenticates with RIME_API_KEY from your environment, or an explicitapi_key.
Fields
str
default:"UNSET"
Reported in metrics only. Prism has a single model and this cannot select another. Defaults to ‘prism’.
Mapping[str, Any] | None
default:"UNSET"
Conversation settings: voice, interruption and how readily a turn ends. Defaults to None (not set).
str | None
default:"UNSET"
WebSocket URL of the Prism endpoint. Defaults to None (not set).