Models Directory

Ultra-fast model optimized for real-time use (~75ms†)

Ultra-fast model optimized for real-time use (~75ms†)

First generation TTS model (outclassed by v2 models)

State-of-the-art multilingual voice designer model (Text to Voice)

First multilingual model (outclassed by v2 models)

Our most lifelike model with rich emotional expression

Human-like and expressive voice design model (Text to Voice)

High quality, low-latency model with a good balance of quality and speed (~250ms-300ms)

High quality, low-latency model with a good balance of quality and speed (~250ms-300ms)

Human-like and expressive speech generation

Hume's Octave text-to-speech system, a speech-language model for expressive and nuanced speech synthesis.