Remy Reference/Models/Text to speech

Text to speech

Models that turn text into spoken audio, with a choice of named voices.

The speech models behind text to speech. Each is priced per million characters and ships a set of named voices you can audition on its page. Some run at real-time latency for live use, others favor quality. Choose a model and a voice to match the tone and speed a moment calls for.

12 models

ElevenLabs TTSText to speech

~$0.20/ 1k characters

Cartesia Sonic 3Text to speech

~$37/ 1M characters

RealtimeLow latency

Gemini 3.1 Flash TTSText to speech

$20.00/ 1M tokens

Audition 30 voices

GPT-4o-mini TTSText to speech

$2.40/ 1M tokens

Audition 9 voices

Grok Voice TTSText to speech

$15/ 1M characters

Audition 28 voices

Kokoro 82MText to speech

$0.62/ 1M chars

Audition 54 voices

Minimax Speech 2.8 HDText to speech

Audition 17 voices

Orpheus 3BText to speech

Canopy Labs· via

$7.00/ 1M chars

Audition 7 voices

Qwen Audio 3.0 TTS PlusText to speech

$27.59/ 1M characters

Audition 2 voices

Qwen3 TTSText to speech

$20.00/ 1M chars

Audition 9 voices

TTS HDText to speech

$30.00/ 1M characters

Audition 6 voices

TTS-1Text to speech

product: {5.00/ 1M characters

Audition 6 voices