ElevenLabsProvider Class

Extends AIProviderBase

Protected Properties

DefaultSupportedTypes : override InferenceType

[Get]

Signature

override InferenceType DefaultSupportedTypes

SynthesizeStreamCoroutine ( text , voice , onReady )

Streams text-to-speech synthesis from ElevenLabs and invokes a callback with a ready AudioClip.

Handles output format selection and decoding based on provider settings.

See ITextToSpeechTask and provider voice settings. Product docs: https://elevenlabs.io/docs/capabilities/text-to-speech⁠

Signature

IEnumerator SynthesizeStreamCoroutine(string text, string voice=null, Action< AudioClip > onReady=null)

Parameters

text: string Input text to synthesize. Empty text short-circuits with a warning.

voice: string Optional voice override; falls back to configured default when null/empty.

onReady: Action< AudioClip > Invoked with the decoded AudioClip when available.

Returns

IEnumerator

TranscribeAsync ( audioBytes , language , ct )

Transcribes speech using ElevenLabs speech-to-text and returns plain text.

Validates audio bytes, attaches optional language hints, and chooses output format.

Signature

async Task< string > TranscribeAsync(byte[] audioBytes, string language=null, CancellationToken ct=default)

Parameters

audioBytes: byte[] Raw audio payload (e.g., WAV/PCM). Must be non-null and non-empty.

language: string Optional BCP-47 code (e.g., "en"). If null, ElevenLabs may auto-detect.

ct: CancellationToken Cancellation token to abort upload or HTTP processing.

Returns

async Task< string > Transcript text extracted from the API response.