Model icon
xai-tts-1
Text to Speech
Model:
xAI TTS 1 is a low-latency multilingual text-to-speech model for natural, expressive voice generation in real-time voice products and content creation.
Input
126/15000

1-15000 characters. Supports inline and wrapping speech tags for expressive delivery.

Only used for MP3 output.

Output

Example Output

This is sample data. Generate a new result to see real output.

Task ID:xai-tts-1-exampleCreated:2026-06-25 22:44:04
Status:finishedProgress:100%
Examples
Pricing details

Transparent pricing with no hidden fees. Pay as you go.

xAIxai-tts-1
Text to speech
PoYo price
$0.012
2.4 credits
Official price
$0.015
Fal
You save
20%

* Actual fees are based on the final output.

Introduction

Complete guide to using Affordable xAI TTS 1 API for Low-Latency Voice

Affordable xAI TTS 1 API for Low-Latency Voice

Build voice agents, support bots, narrations, and multilingual speech products with xAI TTS 1 on PoYo. Use one async API for expressive voices, speech tags, language auto-detection, MP3/WAV/PCM/telephony output, standard task polling, and webhook callbacks with clear per-character billing.

Available xAI TTS 1 API Models on PoYo

01

xai-tts-1

Production-ready text-to-speech at 2.4 credits per 1000 characters, equal to $0.012 per 1000 characters on PoYo, with five voices, language auto-detection, speech tags, and configurable audio output.
Try Playground

Key Features of xAI TTS 1 API

xAI TTS 1 is designed for natural, responsive speech generation in real products, from voice agents to content pipelines.

01

Natural Voice With Low Latency

Generate speech that sounds fluent and expressive without forcing users through a slow batch workflow. This is especially useful for assistants, customer support, and interactive product experiences.

  • Create natural delivery for agents and narration.
  • Use responsive generation for product workflows.
  • Keep billing predictable with per-character pricing.
xAI TTS voice generation

02

Speech Tags For Fine Control

Add expression directly in the text field with inline and wrapping tags. Teams can keep script and performance direction together instead of maintaining a separate voice direction system.

  • Use inline tags such as [laugh], [pause], and [sigh].
  • Use wrapping tags such as <whisper> and <slow>.
  • Guide tone, timing, and emphasis in one text.
xAI TTS speech tags

03

20+ Languages With Auto Detection

Set language_code to auto for mixed or unknown scripts, or choose a BCP-47 code for stricter localization workflows. This helps teams ship multilingual voice output without changing endpoints.

  • Auto-detect language from the text.
  • Support English, Chinese, French, German, Japanese, Korean, Spanish, and more.
  • Use one API shape for multilingual voice products.
xAI TTS multilingual speech

04

Flexible Audio Output Formats

Choose compact MP3, production WAV or PCM, or telephony-oriented mu-law and A-law output. Configure sample rate and MP3 bit rate from the same request payload.

  • Use MP3 for compact playback and delivery.
  • Use WAV or PCM for editing and post-production.
  • Use mu-law or A-law for telephony pipelines.
xAI TTS output formats

What Can You Build With xAI TTS 1 API?

01

Voice Agents

Generate natural assistant responses for product agents, onboarding flows, and real-time support tools.

02

Customer Support Bots

Turn support scripts and dynamic replies into warm, clear speech for automated service experiences.

03

Narration And Voiceovers

Create product demos, short video narration, training audio, announcements, and creator voice tracks.

04

Multilingual Applications

Ship localized spoken output in one workflow with language auto-detection and consistent audio formats.

xAI TTS 1 Pricing: PoYo vs fal.ai

PoYo exposes xAI TTS 1 through the standard async generation workflow with a lower reference unit price based on the provided pricing screenshot.

FeaturePoYoFal.ai
Public model IDxai-tts-1xai/tts/v1
Fal.ai$0.012 per 1000 characters$0.015 per 1000 characters
PoYo credits2.4 credits per 1000 charactersN/A
Max text length15000 characters15000 characters
Output formatsMP3, WAV, PCM, mu-law, A-lawMP3, WAV, PCM, mu-law, A-law

PoYo pricing uses 2.4 credits per 1000 characters. One credit is priced at $0.005.

How to Use xAI TTS 1 API on PoYo

  1. Get your API key: Create a PoYo account and generate an API key from the dashboard. Get API Key ->
  2. Write the script: Enter 1-15000 characters and add tags such as [pause], [laugh], [sigh], <whisper>, or <slow> when you need expressive delivery.
  3. Choose voice and output: Select eve, ara, rex, sal, or leo; set language_code to auto or a supported BCP-47 code; then choose codec, sample rate, and MP3 bit rate.
  4. Submit and retrieve: Submit the task through PoYo's generate API, then poll the standard task status endpoint or receive a webhook callback. View API Docs ->

xAI TTS 1 FAQ

What voices does xAI TTS 1 support?

PoYo exposes the five fal/xAI presets: eve for energetic delivery, ara for warm friendly speech, rex for confident clear speech, sal for smooth balanced output, and leo for authoritative narration.

How much does xAI TTS 1 cost on PoYo?

xAI TTS 1 costs 2.4 credits per 1000 characters on PoYo, equal to $0.012 per 1000 characters. Billing is based on the length of the text field.

What is the maximum text length?

The text field supports 1 to 15000 characters per request. Split longer scripts into multiple tasks if you need chapter-level or long-form audio.

Does xAI TTS 1 support automatic language detection?

Yes. Set language_code to auto to let the model detect the language from the text, or pass a supported BCP-47 code such as en, zh, ja, ko, fr, de, ru, es-MX, or pt-BR.

How do speech tags work?

Speech tags are written directly inside text. Use inline tags such as [laugh], [pause], and [sigh], or wrapping tags such as <whisper>text</whisper> and <slow>text</slow> to guide delivery.

Can I output WAV, PCM, mu-law, or A-law?

Yes. output_format.codec supports mp3, wav, pcm, mulaw, and alaw. You can also set sample_rate, and for MP3 you can set bit_rate.

Is xAI TTS 1 suitable for real-time voice agents?

The model is designed for low-latency natural speech, making it a strong fit for voice agents, support bots, assistant replies, and interactive app workflows.

How do I get the generated audio?

Submit a generation task and use the returned task_id with PoYo's standard status endpoint. If callback_url is provided, PoYo sends a webhook when the task finishes or fails.

Why Use PoYo for xAI TTS 1 API Access

01

Affordable Character Billing

Generate xAI TTS 1 speech at $0.012 per 1000 characters with transparent credit usage.

02

Unified API Workflow

Use the same PoYo submit, status, callback, billing, and file retrieval pattern as other generation models.

03

Production Controls

Control voice, language, codec, sample rate, and bit rate from one developer-friendly payload.

04

Reliable Async Results

Submit tasks, poll stable status, and retrieve audio files through PoYo's standard task output structure.