Interface TTSOptions

Options for AssemblyAI Streaming TTS.

interface TTSOptions {
    apiKey?: string;
    baseUrl?: string;
    language?: null | string;
    model?: string;
    sampleRate?: number;
    textPacing?: boolean;
    tokenizer?: tokenize.SentenceTokenizer;
    voice?: string;
    wordTimestamps?: boolean;
}

Properties

apiKey? baseUrl? language? model? sampleRate? textPacing? tokenizer? voice? wordTimestamps?

Properties

Optional apiKey

apiKey?: string

Defaults to ASSEMBLYAI_API_KEY.

Optional baseUrl

baseUrl?: string

Streaming TTS host, including its wss scheme.

Optional language

language?: null | string

Language code or English language name. Omit to use the voice's default language.

Optional model

model?: string

Model name or alias.

Optional sampleRate

sampleRate?: number

PCM output sample rate.

Optional textPacing

textPacing?: boolean

Pace later sentence requests behind audio already generated for the stream.

Optional tokenizer

tokenizer?: tokenize.SentenceTokenizer

Sentence tokenizer used to split streamed LLM output into requests.

Optional voice

voice?: string

Catalog voice UUID or slug. Defaults to Keegan.

Optional wordTimestamps

wordTimestamps?: boolean

Request and emit word-level timestamps.

Settings

Member Visibility

Theme

On This Page

apiKeybaseUrllanguagemodelsampleRatetextPacingtokenizervoicewordTimestamps