Taigi AI TTS API
Fast, High-Quality, Commercial Use Available

We provide the world's first Taigi AI TTS via API. Easily generate speech with REST API. Supports both Tailo (Taigi Romanization) and Han Characters. Ideal speech synthesis service for business use.

POST https://xxxxxxxx-sample.com/v1/tts
{
  "text": "你好,台灣",
  "model": "model6"
}
{
  "success": true,
  "audio_url": "https://xxxxxxxx.com/audio/abc123.wav"
}

Taigi TTS Technical Strengths

High-quality Taigi speech synthesis API optimized for commercial use.
Please register here.

High-Precision Speech Synthesis

Achieves accurate tone processing and natural pronunciation. The AI faithfully reproduces the unique phonetic features of Taigi.

Easy Connection with REST API

Easily integrate with standard REST API. Add Taigi speech synthesis to your application with just a few lines of code.

Commercial Use License Support

Provides licenses for commercial use. Suitable for apps, games, advertisements, and other business applications.

Stable Server Infrastructure

Stable infrastructure with 99.9% uptime guarantee. Fast audio delivery via global CDN.

Supports Both Tailo and Han Characters

Supports both Taigi Romanization (Tailo) and Han Characters. Choose the optimal input method according to your needs. (Currently, we recommend Han Character input as the model is trained mainly with it.)

Latest AI Model

AI model using the latest neural networks. Reproduces natural intonation and rhythm like a human.

SimpleFlexible Pricing Plans

Choose the best plan according to your usage needs.

*The code examples on this page are for reference only.

Please register here.

Lite 🌱

Trial
Contact us
  • 1000 times/Month
  • Standard voices only
  • Supports WAV format
  • Commercial use not allowed
Register Now

Standard 📚

Popular
Contact us
  • 5,000 times/Month
  • Multiple voice selections
  • Supports WAV format
  • Commercial use allowed
Register Now

Business 🏢

Custom
Contact us
  • Unlimited requests
  • Custom voice creation
  • Priority support
  • Dedicated infrastructure support
Contact Us

Free plan is also available.
Additional requests can be handled. Please contact us for details.

APIDocumentation

Simple API design for integration in minutes

Speech Synthesis Endpoint

POST https://xxxxxxxx-sample.com/v1/tts
{
  "text": "你好,台灣!",  // 台湾語テキスト(漢字または台羅)
  "model": "model5"  
}

Response Example

{
  "success": true,
  "audio_url": "https://xxxxxxxx-sample.com/audio/abc123.wav",
}

Getting Started  Steps

Start using Taigi TTS in 4 easy steps

1

Free Registration

Easily register with your email. Trial available without credit card.

2

API Key Issuance

Issue API key from the dashboard. Start calling the API immediately.

3

Refer to Documentation

4

Ready to Use

DevelopersProfile

佐藤 翔悟 (Shogo Sato)

Expert in Taigi AI Technology. Deeply interested in Taigi language and culture, supporting its promotion through technology.

Development Experience

  • Taigi AI TTS Engine Development (2023-present)
  • Taigi AI TTS App iOS Development (2024-present)
  • Taigi AI TTS App Android Development (2025-present)
Developer Profile

Contact Us

Feel free to contact us for questions about the Taigi AI TTS API, commercial use consultation, or custom development requests.

Your personal information will be managed appropriately according to our privacy policy.

FAQ

Frequently Asked Questions about Taigi AI TTS API

This API supports both Han Characters and Tailo (Taigi Romanization), the main writing systems of Taigi. For Han Character input, the system automatically selects the appropriate pronunciation based on context. For Tailo input, speech synthesis is performed based on accurate phonetic notation. Currently, Han Character input is recommended as the model is mainly trained with it.

Yes, commercial use is allowed with the Pro Plan or higher. It can be used in various commercial applications such as apps, games, advertisements, and educational content. However, the Basic Plan is limited to non-commercial use. Please refer to the terms of service for details on commercial use.

The average response time is less than 500ms. It may vary slightly depending on the length and complexity of the text, but in most cases, the URL of the audio file is returned within 1 second. The audio file itself is delivered via CDN, enabling fast access from anywhere in the world.

It is possible to create custom voice models that reproduce the voices of specific voice actors or characters. (Please obtain permission yourself.) Please contact us via the inquiry form for detailed requirements and pricing.

Currently, WAV format is supported. It provides a lightweight file size. Please contact us if you need higher sound quality. We plan to support other audio formats in the future.

If you have other questions, please feel free to contact us.

Contact Us