​
The basics

LMNT’s AI voice models trained on countless hours of speech data allow you to easily and quickly generate speech that sounds like a real person. You can use LMNT to generate speech for a variety of purposes, including:

  • Voiceovers for videos
  • Audiobooks and storytelling
  • Podcasts
  • Virtual assistants
  • Video game characters
  • Chatbots
  • And more!

​
The voices

We have a number of pre-built voices available for you to use. Each voice is a unique person with their own style and personality. You can see our public voices in the Speech Playground or with a call to our API.

However, our pre-built voices are just the start. You can create your own voices from audio samples that you provide. This allows you to create voices that are unique to your brand or project. Learn more about creating custom voices.

​
Get started

  1. Explore the Speech Playground to see what LMNT can do.
  2. See how to use our REST API or SDKs to generate speech in your application.

​
Other resources

Advanced Synthesis Controls

Fine-tune your speech synthesis with advanced controls.

Low Latency Synthesis

Learn how to optimize your speech synthesis for low latency.

IntroductionAdvanced Controls