Choose Auto, TokenLab Verified, or Official for each request, with prices shown up front. See what's new

ElevenLabs: Eleven v4 Turbo

Generate speech with Eleven v4 Turbo from ElevenLabs. Choose a voice and control delivery with audio tags, stability, similarity settings, and IPA pronunciation.
Compare models
eleven-v4-turbo
AvailableElevenLabsText to speechSync
Input / Output
$40.00 / $0.00
Modalities
Audio

About Eleven v4 Turbo

Eleven v4 Turbo is ElevenLabs' real-time text-to-speech model, the low-latency sibling of Eleven v4. It turns a script into spoken audio in 90+ languages and reads inline audio tags such as whispers or pauses to direct delivery. ElevenLabs aims it at voice agents and live conversation, where quick first audio matters more than the extra polish of the standard v4 model.

Where it works well

  • ElevenLabs cites about 100 ms inference latency and 150 ms to first speech, which is what makes live voice agents feel responsive.
  • Inline audio tags like [whispers] and [excited] shape emotion and delivery, and they replace SSML for most direction.
  • The same voice stays consistent when a line is regenerated, according to ElevenLabs, which keeps agent personas stable across calls.
  • Pronunciation controls cover names, acronyms and technical terms, including IPA, and stability and similarity settings tune how closely a voice is followed.

When to choose another model

  • Eleven v4 is the version built for produced content such as audiobooks and narration, so finished media usually sounds better there.
  • It only speaks text you give it; transcription or understanding of audio needs a separate speech-to-text model.
  • Audio tags steer delivery but do not guarantee an exact result, so scripts that need a specific read should be auditioned first.

Getting started

  1. Create API key

    Create a key in Console, then use it with every model on the platform.

  2. Send your first request

    Copy the example for your language and run it against the endpoint.

    Text to speechTokenLab endpoint
    POST/v1/audio/speech
    curl -X POST "https://api.tokenlab.sh/v1/audio/speech" \
      -H "Authorization: Bearer sk-xxx" \
      -H "Content-Type: application/json" \
      -d '{
      "operation": "tts",
      "model": "eleven-v4-turbo",
      "input": "Hello from TokenLab.",
      "voice": "alloy"
    }'

Parameters

Text to speech

ParameterValues and limitsDefaultNotes
inputNot providedNot providedText length limits: minLength=1, maxLength=5000.
language_codeNot providedNot providedThis field is accepted. The current public contract does not specify a complete value range or list. No additional values are promised. The omitted-field default is not verified.
operationNot providedNot providedNot provided
response_formatmp3_22050_32, mp3_44100_32, mp3_44100_64, mp3_44100_96, mp3_44100_128, mp3_44100_192, pcm_8000, pcm_16000, pcm_22050, pcm_24000, pcm_44100, pcm_48000, ulaw_8000, alaw_8000, opus_48000_32, opus_48000_64, opus_48000_96, opus_48000_128, opus_48000_192mp3_44100_128Not provided
seed≥ 0, ≤ 4294967295Not providedNot provided
voiceNot providedRachelThis field is accepted. The current public contract does not specify a complete value range or list. No additional values are promised.

Fields outside this contract are rejected.

Pricing

The Verified price applies to Verified, which costs less on most models. The Official price is the model maker's published price and applies to the more reliable Official route. Auto bills the route that completes the request.

Text to speech

per 1M characters
Official price
Input $40.00 / Output $0.00
Verified price
—
Discount
—

Usage & activity

Success rate is the share of requests that completed. Latency is how long a full response takes; P95 means 95% of requests finished within that time.

Usage & availability

Last 24 hours
Requests
Success rate
P95 latency
Total tokens
Model performance
Metrics appear once privacy and data volume thresholds are met.

Data is based on aggregate user requests, excluding status checks.

Open in Console

Open Eleven v4 Turbo in Console with a prompt ready to edit or send.

Help me try eleven-v4-turbo with a short audio request at /v1/audio/speech. Show the result and cost.

Use cases

  • Phone and web voice agents

    Speak an LLM's reply as it streams in for customer service, sales or booking assistants where a long pause breaks the conversation.

  • Interactive characters

    Give game or companion characters a steady voice and mark emotion shifts inside a line with tags.

  • Spoken notifications and IVR

    Generate short prompts and confirmations in several languages with a fixed voice and correct pronunciation of product names.

  • Quick script drafts

    Hear a script in near real time before sending it to the slower, higher-polish v4 model for the final recording.

Prompt examples

[warmly] Thanks for calling Brightside Dental. [pause] I can help you book or change an appointment. What works for you?

[whispers] The vault code is [pause] four, seven, nine, two. [excited] And that's the last clue!

Your order number is /ˈɔːrdər/ 48-210. Please read it back so I can confirm we have the right one.

FAQ

What is the difference between Eleven v4 and Eleven v4 Turbo?

Turbo is the real-time variant for voice agents and live conversations, with latency around 100 ms. Eleven v4 is tuned for quality in produced content like audiobooks, podcasts and narration. Both use audio tags and support 90+ languages.

How do audio tags work in Eleven v4 Turbo?

You place natural-language tags in square brackets inside the text, such as [whispers], [excited] or [pause]. The model reads them as direction for how to deliver the next words. ElevenLabs says they are more reliable here than in earlier models.

Which languages does Eleven v4 Turbo speak?

ElevenLabs lists support for 90+ languages with native accents, plus multilingual voice cloning. You pick a voice and send text in the target language; the model handles pronunciation, and you can add IPA for names or terms it gets wrong.

Can I clone a voice and use it with Eleven v4 Turbo?

ElevenLabs states that Professional Voice Clones work with v4 and v4 Turbo across the full emotional range. You supply the voice choice in the request; creating the clone itself happens in your ElevenLabs voice library.

What controls does the request give me?

The catalog describes voice choice, audio tags, stability and similarity settings, and IPA pronunciation. Stability trades expressiveness for consistency, and similarity controls how closely the output follows the chosen voice. Tune them on a sample line before production.

How much does Eleven v4 Turbo cost?

On TokenLab, Eleven v4 Turbo costs Input $40.00 / Output $0.00 per 1M characters. The pricing table above shows the full breakdown. Rates depend on the billing unit, specification, and usage. Compare matching conditions in the model's detailed pricing; a single rate does not determine the total cost.

Which endpoint should Eleven v4 Turbo use?

Use https://api.tokenlab.sh/v1/audio/speech for Eleven v4 Turbo. The request example below shows the matching code shape.

Which operations does Eleven v4 Turbo support?

Eleven v4 Turbo supports Text to speech. Select an operation above to see its endpoint and request example.

Compare Eleven v4 Turbo

Sources

Reviewed Oct 2, 2026

Related models