CosyVoice 3
by FunAudioLLM (Alibaba)
LLM-based zero-shot TTS with voice cloning in 9 languages and 18+ Chinese dialects, plus streaming and instruction control.
Voiceover (text-to-speech) Open source
1. Price: huggingface.co
Open weights, free to download.
huggingface.coNo quote yet: price is read from the official pages.
2. Commercial use: LICENSE on GitHub
“each Contributor hereby grants to You a perpetual, worldwide, non-exclusive, no-charge, royalty-free, irrevocable copyright license to reproduce, prepare Derivative Works of, publicly display, publicly perform, sublicense, and distribute”
Apache-2.0 allows commercial use of the code and, per the model card, the weights.
LICENSE on GitHub3. Credit the author: LICENSE on GitHub
“You must give any other recipients of the Work or Derivative Works a copy of this License”
Speech: CosyVoice 3 by FunAudioLLM (Apache-2.0)
LICENSE on GitHub4. Agent access: github.com/QwenAudio/CosyVoice
There is no package or API. Files are downloaded by hand from the official site.
github.com/QwenAudio/CosyVoiceLicense checked Oct 25, 2026 against 1 official page. Tap a verdict to see its source. How we checked Report an error
CosyVoice 3: open speech model with voice cloning
CosyVoice 3 is a free AI model from Alibaba's FunAudioLLM team that reads text aloud in a natural voice. It can copy a voice from a short sample clip and speaks 9 languages plus many Chinese dialects. You run it on your own computer, and its Apache-2.0 license allows paid products.
Where you can use it
A voice copied from a short clip
Give the model a short recording of a voice and its transcript, and it reads any new text in that voice. Use it for your own voice or one you have permission to use.
Voiceovers in several languages
Make narration for videos, lessons or app tours in Chinese, English, Japanese, Korean, German, Spanish, French, Italian or Russian, even in the same cloned voice.
Tone, speed and dialect on request
Ask in plain words for a dialect such as Cantonese, a faster pace, an emotion or a louder voice, written into the prompt.
Spoken replies that start fast
For a voice assistant, feed text in pieces and play the sound while it is still being made, so the reply starts almost at once.
Fixing hard-to-say words
When a name or term comes out wrong, spell its sound out in the text, with Chinese Pinyin or English phoneme codes.
Your own speech server
Run the model as a small web service in a container on a machine with an NVIDIA card, and call it from your apps.
As reported by FunAudioLLM (Alibaba): Latency as low as 150ms with streaming; 9 common languages and 18+ Chinese dialects or accents; Better than CosyVoice 2.0 in content consistency, speaker similarity and naturalness, as the team reports.
CosyVoice 3 popularity
How to start
Tell your AI assistant
“Set up CosyVoice 3 from its GitHub repository (clone with submodules, Python 3.10), download FunAudioLLM/Fun-CosyVoice3-0.5B-2512 to pretrained_models/, install the requirements from the normal PyPI index, and write a script that reads lines from script.txt and saves one WAV per line in my voice from me.wav with its transcript.”
Show codeStep 1 or Step 2 or Step 3
Get the code with its sub-modules and set up a Python 3.10 environment with conda, as the README shows.
git clone --recursive https://github.com/FunAudioLLM/CosyVoice.git
cd CosyVoice
conda create -n cosyvoice -y python=3.10
conda activate cosyvoice
Download the CosyVoice 3 model files from Hugging Face.
from huggingface_hub import snapshot_download
snapshot_download('FunAudioLLM/Fun-CosyVoice3-0.5B-2512', local_dir='pretrained_models/Fun-CosyVoice3-0.5B')
Load the model with AutoModel and call inference_zero_shot (see the voice cloning recipe above), or run example.py.
python example.py
Mistakes to avoid
The code relies on the Matcha-TTS sub-module, which a plain clone leaves empty.
pip install -r requirements.txt -i https://mirrors.aliyun.com/pypi/simple/ --trusted-host=mirrors.aliyun.com copied as ispip install -r requirements.txt from your normal package index, unless the person chose that mirrorThe README command pulls packages from a third-party mirror and marks it as trusted.
Every CosyVoice 3 example in the docs builds the prompt this way.
The example script says Japanese must be translated to katakana.
Versions before 0.9.0 do not work and the ones in between are not tested.
License
Apache-2.0 license. Free for personal and paid products, and you may change and share it. Keep the license and notices when you pass the code or model files on.
“each Contributor hereby grants to You a perpetual, worldwide, non-exclusive, no-charge, royalty-free, irrevocable copyright license to reproduce, prepare Derivative Works of, publicly display, publicly perform, sublicense, and distribute” LICENSE on GitHub
“You must give any other recipients of the Work or Derivative Works a copy of this License” LICENSE on GitHub
“each Contributor hereby grants to You a perpetual, worldwide, non-exclusive, no-charge, royalty-free, irrevocable copyright license to reproduce, prepare Derivative Works of, publicly display, publicly perform, sublicense, and distribute” LICENSE on GitHub
Passing the files on is allowed with the license; changed files must say that they were changed.
“You must cause any modified files to carry prominent notices stating that You changed the files” LICENSE on GitHub
AllowedWith conditionsNot allowedNot stated or not checked yet
Is CosyVoice 3 free for commercial use? The full answer Not legal advice. Check the license page before you ship.
For your agent
License rules tell your agent what it may do with CosyVoice 3. Paste them into your prompt or AGENTS.md.
- CosyVoice 3 (Voiceover (text-to-speech)). LLM-based zero-shot TTS with voice cloning in 9 languages and 18+ Chinese dialects, plus streaming and instruction control. Price: free; commercial use OK; credit required.
Connect: download from https://huggingface.co/FunAudioLLM/Fun-CosyVoice3-0.5B-2512 (no API).
Rule: Apache-2.0 weights (per the model card) and code; commercial use and products allowed.
Rule: When redistributing the weights or code, include the license and keep notices.
Rule: The GitHub README notes its demo content is for academic purposes only; this covers the samples, not the model.
License: https://github.com/QwenAudio/CosyVoice/blob/main/LICENSE (checked 2026-10-25; details: https://agentgid.com/assets/cosyvoice-3/)
Speech: CosyVoice 3 by FunAudioLLM (Apache-2.0)Alternatives
CosyVoice 3 This page
Pick Chatterbox for an MIT-licensed cloning model with 23 languages and a built-in audio watermark.
Pick OmniVoice when you need far more languages (600+), keeping in mind its tokenizer license limits.
Pick Qwen3-TTS for ready preset voices and voice design from a text description in 10 languages.
Local desktop app for voice cloning, voice design, video dubbing, dictation and transcription in 646 languages, with a local API and an MCP server for agents.
Long-form multi-speaker TTS for podcast-style dialogue, up to 90 minutes and 4 speakers, in English and Chinese.
Questions
Can I use CosyVoice 3 in a paid app?
Yes. The model card lists the Apache-2.0 license, which allows commercial use; keep the license with copies you pass on.
Which languages does it speak?
Chinese, English, Japanese, Korean, German, Spanish, French, Italian and Russian, plus 18 or more Chinese dialects and accents.
Do I need an account to download it?
No. The Hugging Face model is not gated, so the files download without signing in.
Can it speak while it is still generating?
Yes. It supports streaming text in and audio out, which the team reports at latency as low as 150 ms.