---
title: Real-time Text-to-Speech API | Telnyx
canonical: https://telnyx.com/products/text-to-speech-api
---
# Real-time Text-to-Speech API | Telnyx

English

# Real-time Text-to-Speech API

1,300+ voices across Telnyx-hosted, provider, and custom models through one API. Switch voices without rebuilding your application.

[Start building](https://telnyx.com/sign-up)[View pricing](https://telnyx.com/pricing/text-to-speech)

Text-to-speech demo

## Test Text-to-Speech API voices

Enter text, choose a voice, adjust speed, and hear Telnyx text-to-speech output across models in the same workflow.

Loading...

Features

## Production-ready text-to-speech APIs

One TTS API for voice synthesis, streaming, custom voices, and speech controls across all providers.

* 3,900+ voices across 80+ languages  
Access voices across Telnyx, provider, and custom models in 80+ languages. Switch through configuration instead of rebuilding around a new provider.
* Streaming and in-call playback  
Stream audio chunks for real-time responses, generate through REST, or play TTS directly in Voice API and TeXML call flows.
* OpenAI SDK compatibility  
Use the REST TTS API with OpenAI-compatible client patterns when that is the fastest path for your team.
* Custom voice creation and dubbing  
Design synthetic voices, clone from reference audio, or support dubbing workflows in production.
* Pronunciation control  
Use pronunciation dictionaries and voice settings to tune how product names, customer names, and domain terms are spoken.
* Emotion and expression control  
Use SSML, inline speech tags, and emotion markers to direct delivery, tone, and vocal effects across supported providers like xAI Grok, Fish Audio, and Telnyx Ultra.

TTS providers

## Text-to-Speech providers available through Telnyx

Compare Telnyx native voices, xAI Grok TTS, AWS Polly, Azure Speech, ElevenLabs, MiniMax, MurfAI, Resemble AI, Inworld, and Qwen3TTS through one Text-to-Speech API.

[Browse TTS voices](https://developers.telnyx.com/docs/tts-stt/tts-available-voices)

* [![telnyx-native-voices brand logo](https://d2efw2akbx8gaf.cloudfront.net/https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/telnyx-native-voices-1.svg)Telnyx native voicesKokoroTTS, Qwen3TTS, Ultra, and Grok options for live calls, voice AI agents, and REST text-to-speech.Learn more](https://developers.telnyx.com/docs/voice/tts/providers/telnyx)
* [![xai-grok-tts brand logo](https://d2efw2akbx8gaf.cloudfront.net/https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/xai-grok-tts-1.svg)xAI Grok TTSExpressive multilingual voices with speech tags for pauses, laughter, whispering, emphasis, pitch, pace, and intensity.Learn more](https://developers.telnyx.com/docs/voice/tts/providers/xai)
* [![aws-polly brand logo](https://d2efw2akbx8gaf.cloudfront.net/https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/aws-polly-1.svg)AWS PollyUse AWS Polly standard, neural, generative, and long-form voices through Telnyx text-to-speech requests.Learn more](https://developers.telnyx.com/docs/voice/tts/providers/aws)
* [![azure-speech brand logo](https://d2efw2akbx8gaf.cloudfront.net/https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/azure-speech-1.svg)Azure SpeechUse Microsoft Azure neural voices with SSML support and voice format strings through the Telnyx TTS API.Learn more](https://developers.telnyx.com/docs/voice/tts/providers/azure)
* [![elevenlabs brand logo](https://d2efw2akbx8gaf.cloudfront.net/https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/elevenlabs-1.svg)ElevenLabsRelay ElevenLabs voices through Telnyx with provider settings for model ID, language code, stability, style, and speaker boost.Learn more](https://developers.telnyx.com/docs/voice/tts/providers/elevenlabs)
* [![MiniMax brand logo](https://d2efw2akbx8gaf.cloudfront.net/https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/minimax-2.svg)MiniMaxUse streamed MiniMax voices with speed, volume, pitch, language emphasis, and custom voice support.Learn more](https://developers.telnyx.com/docs/voice/tts/providers/minimax)
* [![murfai brand logo](https://d2efw2akbx8gaf.cloudfront.net/https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/murfai-1.svg)MurfAIUse MurfAI Falcon voices with configurable pitch, rate, format, and sample rate through Telnyx provider settings.Learn more](https://developers.telnyx.com/docs/voice/tts/voices/alternative-providers)
* [![resemble-ai brand logo](https://d2efw2akbx8gaf.cloudfront.net/https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/resemble-ai-1.svg)Resemble AIUse Resemble Turbo voices with streamed synthesis, MP3 or WAV output, and precision controls.Learn more](https://developers.telnyx.com/docs/voice/tts/providers/resemble)
* [![inworld brand logo](https://d2efw2akbx8gaf.cloudfront.net/https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/inworld-1.svg)InworldChoose Inworld Mini for speed or Inworld Max for higher quality voices through the Telnyx TTS API.Learn more](https://developers.telnyx.com/docs/voice/tts/providers/inworld)
* [![qwen3tts brand logo](https://d2efw2akbx8gaf.cloudfront.net/https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/qwen3tts-1.svg)Qwen3TTSCreate organization-scoped custom voices in Voice Design Lab and use Qwen3TTS voices through Telnyx.Learn more](https://developers.telnyx.com/docs/voice/tts/providers/telnyx/qwen3)
* [![fish-audio-1 brand logo](https://d2efw2akbx8gaf.cloudfront.net/https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/fish-audio-2.svg)Fish AudioCreate lifelike AI voices with Fish Audio's advanced speech synthesis models through Telnyx.Learn more](https://developers.telnyx.com/docs/voice/tts/providers/fishaudio#provider-docs)

Why Telnyx

## Why use Telnyx for Text-to-Speech API?

Most TTS providers stop at audio generation. Telnyx connects text-to-speech to calls, AI agents, and IVRs while keeping model choice behind one API.

* One API for every voice and provider  
Use one API key, one bill, and one support motion for Telnyx voices, provider models, and custom voices. Switch models through configuration instead of rebuilding your application around each vendor.
* Speech inside calls, agents, and IVRs  
Telnyx connects TTS to Voice API, TeXML, and Voice AI workflows, so generated speech is part of the call flow instead of a separate audio service.
* Run voices on Telnyx infrastructure  
For select Telnyx-run models, speech generation stays on Telnyx infrastructure instead of calling a separate TTS vendor first.
* Pay-as-you-go from your first call  
Start at $3 per million characters with no seat fees, no annual commits, and no separate provider contracts. Pricing stays linear as you scale.
* Global infrastructure and low latency  
Run TTS close to your callers across Telnyx's global anchor sites, reducing round-trip time for real-time voice applications.
* Compliance and data residency  
Store call recordings, transcripts, and AI data in your preferred region, with US, EU, AUS, and UAE data residency available for voice workloads.

Text-to-Speech customer stories

## How teams use Telnyx Voice AI and Text-to-Speech

Read examples from teams using Telnyx for voice AI agents, live calls, contact centers, and global voice apps. These stories show the same requirements that matter for Text-to-Speech API deployments: voice quality, latency, one API, and fewer provider integrations.

* [![Reinvent Telecom logo](https://d2efw2akbx8gaf.cloudfront.net/https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/Reinvent%20Telecom.svg)Reinvent Telecom"The feedback has been very positive, especially around the quality of the AI interaction." David Ansehl Jr., VP of Sales and Marketing. Reinvent built MyCloud AI Receptionist with voice personas and natural pacing.Read story](https://telnyx.com/customer-stories/reinvent-telecom)
* [![Ultradoc logo](https://d2efw2akbx8gaf.cloudfront.net/https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/Ultradoc.svg)Ultradoc"This thing responds so fast." Vincent Sajkowski, CEO and co-founder. Ultradoc chose Telnyx after testing stacks that sent calls across separate vendors.Read story](https://telnyx.com/customer-stories/ultradoc)
* [![Replicant logo](https://d2efw2akbx8gaf.cloudfront.net/https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/Carousel_Thumbnail_Customer-Story_Replicant-2ef6e2ed.jpg?width=1120&format=webp)Replicant"We're able to get latency down to the level we want with Telnyx." Ben Gleitzman, CTO. Replicant uses Telnyx for conversational AI where slow responses make calls feel unnatural.Read story](https://telnyx.com/customer-stories/replicant)
* [![echowin logo](https://d2efw2akbx8gaf.cloudfront.net/https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/Customer-story_Echowin_16x9-59376609.png?width=1120&format=webp)echowin"We wanted a hands-off solution where we didn't have to worry about the base layer." Kaushal Subedi, co-founder and CEO.Read story](https://telnyx.com/customer-stories/echowin)
* [![Ooma logo](https://d2efw2akbx8gaf.cloudfront.net/https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/Carousel_Thumbnail_Customer-Story_Ooma-a6506978.svg)Ooma"We're able to get coverage all over the world through one API and one point of contact." Dayton Turner, VP of Enterprise Product.Read story](https://telnyx.com/customer-stories/ooma)

How it works

## Call the TTS API and receive audio

Send text with a voice ID, then receive generated speech over REST, WebSocket streaming, or a live call flow.

[Read the docs](https://developers.telnyx.com/docs/voice/tts/overview)

Generate speech from text with the TTS API

```
curl --request POST \
  --url https://api.telnyx.com/v2/text-to-speech \
  --header "Authorization: Bearer $TELNYX_API_KEY" \
  --header "Content-Type: application/json" \
  --data '{
    "text": "Your order is ready for pickup.",
    "voice": "Telnyx.Ultra.01eaafa9-308a-4276-a017-6ab0cf061b1f"
  }'
```

**Pricing**

## How much does the Telnyx TTS API cost?

Telnyx TTS starts at $3 per 1 million characters. Pricing varies by model, with volume discounts available for larger deployments.

Starting at

$3.00

Starting cost per 1 million characters

[See pricing](https://telnyx.com/pricing/text-to-speech)[Read the docs](https://developers.telnyx.com/docs/voice/tts/overview)

Products

## TTS on infrastructure built for real-time AI

* [![Voice AI Agents](https://d2efw2akbx8gaf.cloudfront.net/https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/Voice_AI_Agent_2x.webp?width=1024&format=webp)Voice AI AgentsGive production agents a speech layer tuned for turn taking, voice quality, language, and cost.](https://telnyx.com/products/voice-ai-agents)
* [![Voice API and TeXML](https://d2efw2akbx8gaf.cloudfront.net/https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/Voice_API_and_TeXML_2x.webp?width=1024&format=webp)Voice APIPlay generated speech inside call flows without managing a separate voice vendor for every experience.](https://telnyx.com/products/voice-api)
* [![Speech-to-Text](https://d2efw2akbx8gaf.cloudfront.net/https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/Speech-to-Text.webp?width=1024&format=webp)Speech-to-TextPair TTS with transcription for the full listen, reason, and respond loop in one platform.](https://telnyx.com/products/speech-to-text)
* [![Inference](https://d2efw2akbx8gaf.cloudfront.net/https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/inference.webp?width=512&format=webp)InferenceRun open-source LLMs on co-located GPUs in the same private network as your TTS, Voice API, and AI agents. No API middleman between your models and your speech layer.](https://telnyx.com/products/inference)

## Build Text-to-Speech with one API

Start with the voice that fits your first use case, then keep room to switch as quality, speed, language, and cost requirements change.

[Start building](https://telnyx.com/sign-up)[View pricing](https://telnyx.com/pricing/text-to-speech)

## Build Text-to-Speech with one API

Start with the voice that fits your first use case, then keep room to switch as quality, speed, language, and cost requirements change.

[Start building](https://telnyx.com/sign-up)[View pricing](https://telnyx.com/pricing/text-to-speech)

## FAQ

### What is a text-to-speech API gateway?

A text-to-speech API gateway gives developers one API surface for multiple TTS engines, voices, and delivery paths. Telnyx acts as that gateway for Telnyx voices, supported provider models, custom voices, REST generation, WebSocket streaming, and in-call playback.

### Which TTS engines does Telnyx support?

### Which TTS model should I choose?

### Can I use Telnyx TTS in live calls and apps?

### Does Telnyx support low-latency TTS for real-time apps?

### Where is Telnyx TTS API processing hosted?

### Is Telnyx TTS API suitable for compliance-sensitive voice applications?

### Can Telnyx TTS API support model fallback or failover?

### How much does the Telnyx TTS API cost?

### Does Telnyx TTS API support voice cloning and dubbing?

### Does Telnyx TTS API support SSML, prosody, and emotion controls?

### How many voices and languages does Telnyx TTS API support?

### Can Telnyx TTS API handle multilingual voice agents?

### Can I switch between TTS engines without rewriting code?

### How is the Telnyx TTS API different from calling TTS providers directly?

### How does Telnyx TTS API compare to standalone TTS providers and orchestration tools?

### Do I need separate contracts with TTS providers?

[Start building](https://telnyx.com/sign-up)