Skip to content
Docs.Albastov.

Audio models

Text to speech, voice cloning and speech to text, on the same routes with the same key.

Sample 0:08

OmniVoiceomnivoice

Clone a voice, or design one from a description.

Voice cloning600+ languages
Sample 0:09

Kokorokokoro

Fast, clear narration in 54 ready-made voices.

Built-in voices8 languages
Sample 0:14

Parakeetparakeet

Transcripts with punctuation and word timestamps.

Speech to text25 languages

More models 11

Sample 0:08

Qwen3-TTSqwen3-tts

Faithful cloning from 10 s of audio.

Voice cloning10 languages
Sample 0:09

Chatterboxchatterbox

Expressive cloned speech with natural pacing.

Voice cloning19 languages
Sample 0:07

Fish Audio S2 Profish-s2pro

Studio-quality cloned speech.

Voice cloning80+ languages
Sample 0:08

Higgs Audio v3higgs3

Conversational cloned speech in 100+ languages.

Voice cloning100+ languages
Sample 0:09

CosyVoice 3cosyvoice3

Quick cloned speech for short lines.

Voice cloning9 languages
Sample 0:09

IndexTTS 2.5indextts2.5

Cloned speech in English and Chinese.

Voice cloning2 languages
Sample 0:08

BreezeTTS 2breeze2

Cloned speech with a described delivery.

Voice cloning2 languages
Sample 0:09

VoxCPM 2voxcpm2

Cloned speech from a short clip, in 30 languages.

Voice cloning30 languages
Sample 0:08

FireRedTTS 3fireredtts3

Multilingual cloned speech in 24 languages.

Voice cloning24 languages
Sample 0:10

MioTTSmiotts

Light, quick cloned speech.

Voice cloning2 languages
Sample 0:08

MOSS-TTS Localmoss-local

Careful cloning for batch narration.

Voice cloning31 languages