Yinyue AI Music
Say it as a song
Creation runs in two steps: the model drafts lyrics first at no cost, you edit or ask for a rewrite in the same screen, and only then is the song generated from a mix of 12 moods and 13 styles, up to four takes per run for comparison. Voice input works in the browser or in the app and uses an OpenAI-compatible speech protocol, swappable for a self-hosted transcriber. MVs come out in 1080x1920 vertical or 1920x1080 horizontal, with lyrics highlighted line by line against the real structure (intro, verse, chorus, outro) and an even split fallback when no timing file exists.
The platform ships a community (trending, new, liked, favourites, history, comments, profile, share text) and an incentive layer (signup, check-in, daily tasks, referral codes, points and balance accounts, with payment callbacks ready). The admin console covers statistics, users and points, orders, task monitoring, notices and feature switches that hide capabilities across clients. Under load, the queue absorbs peaks, failures refund automatically, and Redis rate limits plus account locks prevent over-deduction.
Core capabilities
- Free lyric draft before anything is charged
- 12 moods x 13 styles
- 1-4 takes per run for comparison
- Voice-to-lyrics input, swappable engine
- Vertical and horizontal MVs
- Line-level LRC highlighting on real song structure
- Community, charts, favourites and sharing
- Nine admin feature switches across all clients
- Queue-based peak shaving with automatic refund
Who uses it
- Background music for short-video creators
- Ambient tracks for stores and campaigns
- Event and brand theme songs
- Fast demos for independent musicians
Delivery models
SaaS · on-premises · source licence · bring-your-own engine
Tech stack
Go + Gin + GORM + JWT · Vue3 + Element Plus · Electron · uni-app · MySQL / Redis / RabbitMQ