@squidcloud/client
    Preparing search index...

    Interface AnswerOpenAiVoiceCallRequest

    Request to answer a phone call that reached an OpenAI project's SIP endpoint. Squid accepts the call with the given instructions, and for the call's life runs the agent whenever the realtime model asks it something and hangs up when the model ends the call. Requires the application API key.

    interface AnswerOpenAiVoiceCallRequest {
        agentId: string;
        allowInterruptions?: boolean;
        callId: string;
        chatOptions?: AllAiAgentChatOptions;
        externalCallId?: string;
        greeting?: string;
        greetingMode?: "phrase" | "agent";
        instructions: string;
        integrationId: string;
        interruptionSensitivity?: "low" | "high" | "normal";
        metadata?: Record<string, unknown>;
        model?: string;
        voice?: string;
    }
    Index
    agentId: string

    The agent the realtime model asks for knowledge and actions.

    allowInterruptions?: boolean

    Whether caller speech interrupts the realtime model. Omitted values keep OpenAI's default.

    callId: string

    The call id from OpenAI's realtime.call.incoming webhook.

    chatOptions?: AllAiAgentChatOptions

    Chat options applied to every turn the agent runs for the call: instructions, agent context.

    externalCallId?: string

    The telephony provider's id of the call (a Twilio CallSid), echoed on every session event.

    greeting?: string

    Spoken as soon as the call is answered; without it the model waits for the caller to speak first.

    greetingMode?: "phrase" | "agent"

    agent has the agent compose the opening line, with greeting as its fallback; phrase (the default) speaks greeting as is.

    instructions: string

    Instructions of the realtime model.

    integrationId: string

    The openai_voice integration whose API key answers the call.

    interruptionSensitivity?: "low" | "high" | "normal"

    How readily OpenAI detects caller speech. Omitted values keep OpenAI's default.

    metadata?: Record<string, unknown>

    Free-form data echoed on every session event, for the caller's own bookkeeping.

    model?: string

    Realtime model that talks on the call; defaults to the integration's default model.

    voice?: string

    Voice of the realtime model; defaults to the integration's default voice.