Rime’s TTS functionality exposes both a large set of demographically diverse set of voices to users as well as an increasingly large toolkit for generating voices on the fly.

​
Demographic categories

Rime currently has many voices across a number of demographic categories (including Southern, African American, Latina, Midwestern, and LGBT voices) as well as age ranges from young to old.

​
TTS and Genre

Our customers have many use cases for TTS and we know that different uses require different TTS performances, which we also call genre. Right now we have models for the following genres:

General: A catch-all use case. Works decently for everything.

Conversational: Sounds like normal people having a natural conversation with a dependable range of emotions. When natural and human qualities are the goal, these are the voices to use.

Narration: Works great for narrating books, articles, or newsreaders.

IVR: Perfect for interactive voice agents, call centers, and food ordering. Humanlike performance with the natural cadence of a seasoned call center agent. Works great with filler words. Can have a conversational or more professional style depending on the voice parameter, and captures appropriate emotions based on callflow-specific text.

​
Using the Dashboard

Your Dashboard is a great place to find a voice. By specifying filters, you can find Rime voices that meet your needs.

dashboard-view

Using these filters, and using the play button to hear the voices, you’ve determined Colby will be a great voice for your application. You can then use Colby as the speaker parameter for Rime’s TTS API.

Note: after you get an account you can use the dashboard to listen to audio samples and find the right voice for your application. To generate audio you will need an active API key accesible and an active subscription.

IntroductionLinguistics and TTS