Open source repo

What's inside QwenAudio/CosyVoice

QwenAudio/CosyVoice is a Python repo with 21,889 lines written in 71 files. CosyVoice is built with FastAPI and uvicorn.

Its description on GitHub: “Multi-lingual large voice generation model, providing inference, training and deployment full-stack ability.”

PythonApache-2.024,279

Read on November 4, 2026, at commit 074ca6d.

What the card proved: what it’s built with on the left, what it uses on the right.

See each one with its proof

More on this repo

TechnologiesWhat is CosyVoice built with?CosyVoice is built with FastAPI and uvicorn.

+1 Who writes itHow much of CosyVoice did AI write?0%No AI tool signed any of the 415 commits of CosyVoice.

In figures

lines written: 21,889; The ones people and agents wrote: product, tests, schemas and tooling. Generated code doesn’t count.
files: 71; The product’s, not counting tests.
test files: 1; The files that test the code.
technologies: 9; Each service with the file and line that prove it.

SeeRepo read the facts in its code and kept none of it. AI only helps sort its packages and folders by their names, and never sees its code.

Frequently asked questions

What is QwenAudio/CosyVoice?

QwenAudio/CosyVoice is a Python repo with 21,889 lines written in 71 files. Its description on GitHub: “Multi-lingual large voice generation model, providing inference, training and deployment full-stack ability.”

The proof: its page on GitHub

What is CosyVoice built with?

CosyVoice is built with FastAPI and uvicorn.

The proof: each service with its file and line

What license does CosyVoice have?

CosyVoice is released under the Apache-2.0 license.

The proof: its page on GitHub

How many stars does CosyVoice have on GitHub?

CosyVoice had 24,279 stars on GitHub on November 4, 2026.

The proof: its page on GitHub