ElevenLabs

Configuration#

'elevenlabs' => [
    'api_key' => env('ELEVENLABS_API_KEY', ''),
    'url' => env('ELEVENLABS_URL', 'https://api.elevenlabs.io/v1/'),
]

Speech-to-Text#

ElevenLabs provides speech-to-text through their Scribe model with support for diarization and audio event tagging.

Basic Usage#

use Prism\Prism\Facades\Prism;
use Prism\Prism\ValueObjects\Media\Audio;

$audioFile = Audio::fromLocalPath('/path/to/recording.mp3');

$response = Prism::audio()
    ->using('elevenlabs', 'scribe_v1')
    ->withInput($audioFile)
    ->asText();

Provider-specific Options#

Language Detection#

$response = Prism::audio()
    ->using('elevenlabs', 'scribe_v1')
    ->withInput($audioFile)
    ->withProviderOptions([
        'language_code' => 'en',
    ])
    ->asText();

Speaker Diarization#

$response = Prism::audio()
    ->using('elevenlabs', 'scribe_v1')
    ->withInput($audioFile)
    ->withProviderOptions([
        'diarize' => true,
        'num_speakers' => 2,
    ])
    ->asText();

Audio Event Tagging#

$response = Prism::audio()
    ->using('elevenlabs', 'scribe_v1')
    ->withInput($audioFile)
    ->withProviderOptions([
        'tag_audio_events' => true,
    ])
    ->asText();

Text-to-Speech#

use Prism\Prism\Facades\Prism;

$response = Prism::audio()
    ->using('elevenlabs', 'eleven_multilingual_v2')
    ->withVoice('your-voice-id')
    ->withInput('Welcome to the application.')
    ->withProviderOptions([
        'model_id' => 'eleven_multilingual_v2',
        'speed' => 1.0,
        'voice_settings' => ['stability' => 0.5, 'similarity_boost' => 0.75],
    ])
    ->asAudio();

file_put_contents('speech.mp3', $response->audio->rawContent());

withVoice() selects the voice ID in the endpoint. Set the TTS model through the model_id provider option: the model passed to using() is not mapped to this field. Speech-to-text uses the model from using() normally. Choose a model and voice available in your ElevenLabs account.

TTS options include:

Option	Use
`model_id`	Synthesis model
`voice_settings`	`stability`, `use_speaker_boost`, `similarity_boost`, `style`
`speed`	Mapped into `voice_settings.speed`
`language_code`, `seed`	Language and generation seed
`previous_text`, `next_text`	Adjacent text context
`previous_request_ids`, `next_request_ids`	Adjacent generation context
`apply_text_normalization`, `apply_language_text_normalization`	Text normalization controls

Other provider options are forwarded as payload fields. A failed TTS HTTP response throws a generic \Exception; catching only PrismException does not catch it. Speech-to-text HTTP failures use Laravel's RequestException.

Limitations#

This provider implements audio generation and transcription. Text generation, custom tools and embeddings are unsupported. See Audio for shared response handling.