GOD.EXEA god's-eye view of the AI universe
2026Drag to orbit · right-drag to pan · scroll to zoom
GOD.EXE is booting…

Model

CosyVoice

CosyVoice is an open-weight AI model from Alibaba 阿里巴巴, released in 2024, made for developers.

A multilingual speech generation model from Alibaba for text-to-speech and zero-shot voice cloning.

Main jobs: Text to speech, Voice cloning. It is open source and free to use. You can use it on your own server, offline on your own machine and through an API.

Visit github.com

Markdown version of https://godooo.ai/en/model/cosyvoice. Every page on this site has one: add .md to its address.

At a glance

Type
Model
Released
2024
Open source
Yes · Apache-2.0
GitHub stars
23,759
How you run it
Self-host · Runs offline · API
Made for
Developers
Website
github.com

What it can do

  • Text to speechmain jobunconfirmed

    “Fun-CosyVoice 3.0 is an advanced text-to-speech (TTS) system based on large language models (LLM)”— github.com, 2026-09-24
  • Voice cloningmain jobunconfirmed

    “supports both multi-lingual/cross-lingual zero-shot voice cloning”— github.com, 2026-09-24

unconfirmedproposed by a machine, awaiting calibration

Alternatives

Connections

Questions

What is CosyVoice?

CosyVoice is an open-weight AI model from Alibaba 阿里巴巴, released in 2024, made for developers. A multilingual speech generation model from Alibaba for text-to-speech and zero-shot voice cloning.

Is CosyVoice free?

Yes. CosyVoice is open source (Apache-2.0) and free to use.

Is CosyVoice open source?

Yes, under the Apache-2.0 license. The source code is at https://github.com/QwenAudio/CosyVoice.

Can CosyVoice run locally?

Yes. CosyVoice can run offline on your own machine. It can also be self-hosted on your own server.

What can CosyVoice do?

Main jobs: Text to speech, Voice cloning.

Who makes CosyVoice?

CosyVoice is made by Alibaba 阿里巴巴.

What are open-source alternatives to CosyVoice?

Open-source ones: Chatterbox, F5-TTS, Fish Speech, GPT-SoVITS, ChatTTS, RVC and MoneyPrinterTurbo. Others: ElevenLabs, CapCut 剪映, Descript, WaveNet and HeyGen.

Sources