Interactive demo

Voice synthesis and customization

Everything you need to get started

Natural-sounding neural voices

Benefit from our deep neural networks trained on human speech to automatically produce smooth and natural sounding voice quality.

Custom voices

Design your own unique branded neural voice modeled after your chosen speaker using as little as one hour of recordings. Premium feature.

Controllable speech attributes

Easily adjust pronunciation, volume, pitch, speed and other attributes using Speech Synthesis Markup Language.

Customized word pronunciations

Clarify the pronunciation of unusual words with the help of IPA or the IBM SPR.


Control tone of voice by choosing a specific speaking style: GoodNews, Apology, and Uncertainty.

Voice transformation

Personalize voice quality by specifying attributes such as strength, pitch, breathiness, rate, timbre, and more.

Get started now with IBM Watson Text to Speech