Skip to main content

AudioChunk

A single audio chunk from streaming:

AudioResponse

Complete audio result from generate():

WordTimestamp

Word-level time alignment:

Model

VoiceListResponse

Paginated response from voices().list():

Voice

The current Java Voice/VoiceDetail models still bind legacy language and embedded references fields. The API now returns supported_languages and exposes references through voices().listReferences(...), so getLanguage() is normally null and VoiceDetail.getReferences() is normally empty. VoiceReference.getFilename() likewise binds a legacy field; use getId() and getReferenceText(), or the raw Voice API when you need the current name, path, and URL fields.

Audio Utilities

The AudioFormats utility class provides codec helpers:

Complete Example


Back to Quickstart — or see the API Reference for the raw HTTP contract.