Class RealtimeModel

Parameters

• options: {
      additionalLanguages?: string[];
      additionalParams?: Record<string, unknown>;
      apiKey?: string;
      assistantBackchannelAggressiveness?: number;
      audioSpeed?: number;
      baseUrl?: string;
      boostedKeywords?: string[];
      configsForTools?: PhonicToolConfig[];
      configurationEndpoint?: null | ConfigurationEndpoint;
      connOptions?: APIConnectOptions;
      defaultLanguage?: string;
      enableAssistantBackchannel?: boolean;
      enableRedaction?: boolean;
      forbidSpeechAfterToolCall?: string[];
      generateNoInputPokeText?: boolean;
      generateWelcomeMessage?: boolean;
      intelligenceLevel?: IntelligenceLevel;
      isWelcomeMessageInterruptible?: boolean;
      languages?: string[];
      mcpServers?: string[];
      minWordsToInterrupt?: number;
      model?: string;
      multilingualMode?: "auto" | "request";
      noInputEndConversationSec?: number;
      noInputPokeSec?: number;
      noInputPokeText?: string;
      observabilityIntegrations?: "braintrust"[];
      onConversationCreated?: ((conversationId) => void);
      phonicAgent?: string;
      phonicTools?: string[];
      project?: string;
      pronunciationDictionary?: Item[];
      templateVariables?: Record<string, string>;
      vadMinSilenceDurationMs?: number;
      vadMinSpeechDurationMs?: number;
      vadPrebufferDurationMs?: number;
      vadThreshold?: number;
      voice?: string;
      websocketTimeoutSec?: number;
      welcomeMessage?: string;
  } = {}

  • Optional additionalLanguages?: string[]

    Further ISO 639-1 codes (must not include defaultLanguage)

  • Optional additionalParams?: Record<string, unknown>

    Additional runtime parameters forwarded to Phonic

  • Optional apiKey?: string

    Phonic API key. If not provided, will attempt to read from PHONIC_API_KEY environment variable

  • Optional assistantBackchannelAggressiveness?: number

    How aggressively the assistant backchannels. Only applies when enableAssistantBackchannel is true

  • Optional audioSpeed?: number

    Audio playback speed

  • Optional baseUrl?: string

  • Optional boostedKeywords?: string[]

    Keywords to boost in speech recognition

  • Optional configsForTools?: PhonicToolConfig[]

    Per-tool behavior overrides, one PhonicToolConfig per tool (keyed by name); omitted fields fall back to the plugin defaults. See the README for the available fields.

  • Optional configurationEndpoint?: null | ConfigurationEndpoint

    Endpoint the agent calls to fetch per-conversation configuration. Pass null to disable

  • Optional connOptions?: APIConnectOptions

    Connection options for the API connection

  • Optional defaultLanguage?: string

    ISO 639-1 default language for recognition and speech

  • Optional enableAssistantBackchannel?: boolean

    When true, the assistant produces backchannel responses (e.g. "mm-hmm") while the user speaks

  • Optional enableRedaction?: boolean

    When true, PII/PHI is redacted from transcripts and bleeped from audio after the conversation ends

  • Optional forbidSpeechAfterToolCall?: string[]

    Deprecated

    Use configsForTools with forbid_speech_after_tool_call per tool instead. When set, each listed tool is merged into configsForTools as forbid_speech_after_tool_call: true (an explicit configsForTools entry wins).

  • Optional generateNoInputPokeText?: boolean

    Auto-generate poke text when user is silent

  • Optional generateWelcomeMessage?: boolean

    When true, the welcome message will be automatically generated and welcomeMessage will be ignored

  • Optional intelligenceLevel?: IntelligenceLevel

    LLM intelligence level, standard or high

  • Optional isWelcomeMessageInterruptible?: boolean

    When false, the welcome message cannot be interrupted by the user

  • Optional languages?: string[]

    Deprecated. Use defaultLanguage and additionalLanguages instead. When both of those are omitted and this is set, languages[0] is the default language and languages.slice(1) are additional languages.

  • Optional mcpServers?: string[]

    Names of pre-configured MCP servers to make available to the assistant. Names must be unique

  • Optional minWordsToInterrupt?: number

    Minimum number of user words required to interrupt the assistant

  • Optional model?: string

    The name of the model to use. Defaults to 'merritt'

  • Optional multilingualMode?: "auto" | "request"

    auto: detect language per utterance. request: switch only when the user asks (recommended).

  • Optional noInputEndConversationSec?: number

    Seconds of silence before ending conversation

  • Optional noInputPokeSec?: number

    Seconds of silence before sending poke message

  • Optional noInputPokeText?: string

    Poke message text (ignored when generateNoInputPokeText is true)

  • Optional observabilityIntegrations?: "braintrust"[]

    Observability integrations to forward traces to (currently braintrust)

  • Optional onConversationCreated?: ((conversationId) => void)

    Called with the Phonic conversation ID once the conversation is created

    • • (conversationId): void

      • Parameters

        • conversationId: string

        Returns void

  • Optional phonicAgent?: string

    Phonic agent to use for the conversation. Options explicitly set here will override the agent settings.

  • Optional phonicTools?: string[]

    Phonic tool names available to the assistant

  • Optional project?: string

    Project name to use for the conversation. Defaults to main

  • Optional pronunciationDictionary?: Item[]

    List of { word, pronunciation } entries; words must be unique

  • Optional templateVariables?: Record<string, string>

    Variables substituted into the system prompt and welcome message

  • Optional vadMinSilenceDurationMs?: number

    Minimum silence duration for VAD, in milliseconds

  • Optional vadMinSpeechDurationMs?: number

    Minimum speech duration for VAD, in milliseconds

  • Optional vadPrebufferDurationMs?: number

    Voice-activity-detection prebuffer duration, in milliseconds

  • Optional vadThreshold?: number

    Voice-activity-detection threshold

  • Optional voice?: string

    Voice ID for agent outputs

  • Optional websocketTimeoutSec?: number

    Seconds of inactivity before the Phonic websocket is closed

  • Optional welcomeMessage?: string

    Welcome message for the agent to say when the conversation starts. Ignored when generateWelcomeMessage is true