ElevenLabs

ElevenLabs Speech

ElevenLabs Speech is a text-to-speech workflow for narration, dialogue, voice-over, and other spoken content. Choose a voice and model, then tune delivery with language, stability, similarity, style, speed, and output-format controls exposed by Omnipix. Always review the complete recording for pronunciation, pacing, and unwanted spoken instructions before publishing it.

How to use this model

  1. 1

    Paste the final script and remove notes or directions that should not be spoken.

  2. 2

    Choose a voice and the synthesis model that fits the language and delivery you need.

  3. 3

    Set the language code when useful, then adjust stability, similarity, style, and speed conservatively.

  4. 4

    Select an output format that matches the next editing, playback, or telephony step.

  5. 5

    Generate and listen from beginning to end; revise the text or settings where pronunciation and pacing need correction.

Text to SpeechAvailable

Text to Speech

This version exposes the ElevenLabs text-to-speech workflow directly: supply text, a voice ID, a synthesis model, and an output format. Optional language and voice settings control stability, similarity, style, speaker boost, and speed. The text itself carries much of the intended delivery, so punctuation and wording should be treated as performance controls.

Try now

Properties

  • Requires text, a voice ID, a synthesis model, and an output format.
  • Supports an optional language code and detailed voice settings.
  • Offers multiple speech-oriented audio encodings through the format selector.

How to use this model

  1. 1

    Enter the final text without production notes that should remain silent.

  2. 2

    Select the voice ID, synthesis model, and output format.

  3. 3

    Set a language code when it helps disambiguate the script.

  4. 4

    Tune stability, similarity, style, speaker boost, and speed in small increments.

  5. 5

    Generate and listen for pronunciation, pacing, clipping, and accidentally spoken directions.

Examples

Public text-to-speech output generated in Omnipix.

Parameters

Text

text
Required

Text to synthesize.

Voice Id

voice_id
Required

ElevenLabs voice ID.

Model Id

model_id
Required

ElevenLabs TTS model ID.

Default: eleven_multilingual_v2

  • Flash v2.5

  • Turbo v2.5

  • Multilingual v2

  • v3

Output Format

output_format
Required

ElevenLabs output format, passed as the output_format query parameter.

Default: mp3_44100_128

  • mp3_22050_32

  • mp3_24000_48

  • mp3_44100_32

  • mp3_44100_64

  • mp3_44100_96

  • mp3_44100_128

  • mp3_44100_192

  • pcm_8000

  • pcm_16000

  • pcm_22050

  • pcm_24000

  • pcm_32000

  • pcm_44100

  • pcm_48000

  • ulaw_8000

  • alaw_8000

  • wav_8000

  • wav_16000

  • wav_22050

  • wav_24000

  • wav_32000

  • wav_44100

  • wav_48000

  • opus_48000_32

  • opus_48000_64

  • opus_48000_96

  • opus_48000_128

  • opus_48000_192

Language Code

language_code
Optional

Optional ISO language code for multilingual models.

ElevenLabsVoiceSettings

voice_settings
Optional