Introduction

Rime is a new text-to-speech tool that is incredibly flexible and allows you to optimize for the voices that your users hear. Currently Rime's TTS technology can be reached via REST API, but we are rapidly iterating and adding features, including client libraries, in the very near future.

Features

• Hundreds of voices to choose from, each distinctive and unique.
• Manipulate nuanced details of the created speech including punctuation, pauses, and pronunciations.
• Conversational tone and context-responsive generation of unique linguistic elements including exclamations and filled pauses like um, uh, like, etc.
• Extremely low latency for much-faster than realtime generation for any possible use case.

How does it work?

Rime lets you synthesize speech from an extremely diverse set of voices. Rime has trained extremely large speech models on a large proprietary dataset to create the most flexible and most interesting voices around.