Skip to main content

​
Welcome to MusicMint

MusicMint gives you a single, consistent REST API to generate professional-quality music, lyrics, audio stems, and more — powered by multiple best-in-class AI models under the hood. Whether you’re building a music app, a content creation tool, or automating audio production at scale, MusicMint handles the complexity so you can focus on your product.

Quick Start

Get your first track generated in under 5 minutes.

API Reference

Explore all endpoints, parameters, and response formats.

Authentication

Create and manage your AppKey.

Webhooks

Get notified when long-running tasks complete.

​
Core capabilities

​
Music generation

Create original, full-length tracks from a text prompt or style tags. Specify mood, genre, instrumentation, or paste in your own lyrics. MusicMint routes your request to the best available model and returns a stream-ready MP3.

​
Cover & remix

Transform an existing clip with new style, key, or lyrics. Use the remaster endpoint to clean up or re-process audio without changing the underlying composition.

​
Vocal tools

  • Add Vocals — layer AI-generated vocals on top of an instrumental
  • Add Instrumental — add a backing track beneath an uploaded vocal recording
  • Sample / Humming — hum a melody, upload it, and get a fully produced song back

​
Lyrics generation

Generate song lyrics from a prompt or topic using large language models. Returns structured verses, chorus, and bridge sections ready to paste into a generation request.

​
Audio processing

  • Stem separation — split a track into vocals, drums, bass, and other
  • MIDI export — extract a MIDI representation of any clip
  • Waveform / WAV — get raw waveform data or lossless audio
  • Lyrics timeline — per-word timestamps for karaoke or subtitle sync

​
AI Singer (Persona)

Train a custom AI voice model from a short audio sample. Once created, the persona can sing any lyrics in a consistent, recognizable voice.

​
How it works

All generation endpoints are asynchronous. You submit a request, receive a task_id, then either poll GET /api/v1/task/{id} or receive a webhook callback when the result is ready. Most tracks complete in 30–120 seconds. A few utility endpoints (audio upload, concat) are synchronous and return results immediately.

​
Reliability & scale

  • Multi-model routing — requests are distributed across providers with automatic failover
  • Priority queues — your API tier determines queue priority; Pro keys skip the free queue
  • TTL-managed storage — generated files are stored on CDN with configurable retention (default 7 days)
  • Signed webhooks — every callback is signed with HMAC-SHA256 so you can verify authenticity

​
Rate limits & quotas

Limits are enforced per AppKey and configured in the dashboard. Contact support to request higher limits for production workloads.