YuE
Open-source model that turns lyrics into full songsGenerate complete songs with vocals and accompaniment from lyrics
Highlights
What you can do with it
- Generate a pop, metal, or jazz song from written lyrics
- Experiment with multilingual and cross-language songs
- Research lyric-to-song generation and vocal modeling
How to use it
- 1 Download the model checkpoints from the YuE repository.
- 2 Prepare lyrics and pick a language and style.
- 3 Split the lyrics into segments with the provided tools.
- 4 Generate music for each segment, then stitch them together.
- 5 Optionally run an upsampler and export the final track.
Who it is for
Pros
- Free and open-source to use and study
- Generates full songs with vocals and accompaniment
- Supports many languages and styles
- Provides checkpoints and upsamplers for flexibility
Limitations
- Technical setup with a real learning curve
- Needs capable hardware to run well
- Open-source terms and model updates can change
- Output quality varies and often needs manual editing
What YuE is
YuE is an open-source music generation model. It can generate full songs up to five minutes long, including vocals and accompaniment, based on given lyrics. The model supports multiple languages such as English, Chinese, Japanese, and Korean, and styles such as pop, rock, jazz, and rap. It is currently free and open-source, and it provides checkpoints and upsamplers so users can expand and adjust it. The project is aimed at researchers and creators exploring lyric-to-song generation. Because it is a technical model rather than a simple web app, there is a learning curve, and running it well depends on your own hardware and setup.
What you can do with it
- Turn written lyrics into a full song with vocals and accompaniment.
- Generate across languages such as English, Chinese, Japanese, and Korean.
- Try styles such as pop, rock, jazz, and rap.
- Use checkpoints and upsamplers to expand and refine output.
Who it is for
- Researchers studying music generation models.
- Musicians exploring lyric-to-song generation.
- Developers experimenting with open-source audio models.
- Creators comfortable with a technical, hands-on setup.
What to watch out for
YuE is a technical model, so expect a learning curve and the need for suitable hardware to run it well. Output quality varies, and generated songs often need manual editing and stitching to sound finished. Because it is open-source, terms, checkpoints, and updates can change, so follow the project repository for current details. Treat it as a research and creative tool rather than a one-click product, and plan time to prepare lyrics and process segments. Run one full task end to end first, then expand once you are comfortable with the workflow and the result quality.
FAQ
Is YuE free?
Yes. The model is currently free and open-source.
What can it generate?
It can generate full songs up to five minutes with vocals and accompaniment from lyrics.
Which languages and styles?
It supports languages such as English, Chinese, Japanese, and Korean, and styles such as pop, rock, jazz, and rap.
Is it easy to use?
It is a technical model, so expect a learning curve and the need for suitable hardware.
Why it is free
What the official page says： 免费
Free limits, trial length and features are decided by the official site and can change at any time.
Product preview
Screenshot archived from the listing page, for preview only.
