Back to Nodes
Minimax Music 01

Minimax Music 01

Official

AI music generation model that synthesizes music with lyrics and vocals based on a reference track.

Nodespell AI
AI / Audio / Minimax

AI music generation model that synthesizes music with lyrics and vocals based on a reference track.

Model Overview

A powerful AI music generator that creates up to 60 seconds of music, complete with lyrics and vocals, in the style of a reference track you provide. It's designed to be intuitive for creatives, allowing you to guide the musical output with your own lyrics and sonic inspiration.

Best At

  • Style transfer: Mimicking the musical style, instrumentation, and vocal qualities of a reference song.
  • Lyric-to-music: Generating vocal melodies and accompanying music directly from your provided lyrics.
  • Quick demos: Rapidly creating short musical pieces for soundtracks, AI singer compositions, or musical reinterpretations.

Limitations / Not Good At

  • Output length: Currently limited to 60 seconds of audio. (3 minutes planned for the next release).
  • Reference required: Needs a reference track (song, voice, or instrumental) to learn the desired style.
  • Lyric length: Maximum of 350-400 characters for lyrics.

Ideal Use Cases

  • Creating background music for videos or podcasts.
  • Experimenting with different musical styles for song ideas.
  • Generating unique vocal tracks for AI singer projects.
  • Producing quick musical sketches or demos.

Input & Output Format

Input:

  • lyrics (string, optional): Your song lyrics, with newlines for line breaks and pauses.
  • song_file (audio, optional): A reference song (.wav or .mp3, >15s) for overall style.
  • voice_file (audio, optional): A reference voice (.wav or .mp3, >15s) for vocal style.
  • instrumental_file (audio, optional): A reference instrumental (.wav or .mp3, >15s) for accompaniment style.
  • voice_id (string, optional): Reuse a previously uploaded voice.
  • instrumental_id (string, optional): Reuse a previously uploaded instrumental.
  • sample_rate (integer, optional): Desired output sample rate.
  • bitrate (integer, optional): Desired output bitrate.

Output:

  • A URI (string) pointing to the generated music file (MP3).

Performance Notes

  • Generates up to 60 seconds of music quickly.
  • Style learning is based on the provided reference track(s).
  • Handles multiple genres including classical, pop, rock, and electronic.

Model Examples (3)

Example Index01 / 03
Example 01

Mystery-thriller song from cue reference

Reference-based vocal song built from an existing tension cue.

Open
Source Inputs03
Lyrics

[Verse] Paper dust, fluorescent blue Half the city never knew We kept the copies in the dark Waiting for the first good spark [Chorus] Truth moves slow through midnight rooms But it still burns through all the gloom If the signal holds, if the wires survive Morning finds us still alive

Voice File
Example input
Instrumental File
Example input
Parameters03
Lyrics
[Verse]
Paper dust, fluorescent blue
Half the city never knew
We kept the copies in the dark
Waiting for the first good spark

[Chorus]
Truth moves slow through midnight rooms
But it still burns through all the gloom
If the signal holds, if the wires survive
Morning finds us still alive
Sample Rate
44100
Bitrate
256000
musicreference-audiosong
Response
Inputs (4)

Lyrics

String

Lyrics. Use to separate lines. Supports [intro][verse][chorus][bridge][outro]. Valid input: 10-600 characters.

Multi InputMin: 0Max: 100

Song File

String

Reference song, should contain music and vocals. Must be a .wav or .mp3 file longer than 15 seconds.

Min: 0Max: 100

Voice File

String

Voice reference. Must be a .wav or .mp3 file longer than 15 seconds. If only a voice reference is given, an a cappella vocal hum will be generated.

Min: 0Max: 100

Instrumental File

String

Instrumental reference. Must be a .wav or .mp3 file longer than 15 seconds. If only an instrumental reference is given, a track without vocals will be generated.

Min: 0Max: 100
Parameters (5)

Lyrics

String

Lyrics with optional formatting. You can use a newline to separate each line of lyrics. You can use two newlines to add a pause between lines. You can use double hash marks (##) at the beginning and end of the lyrics to add accompaniment. Maximum 350 to 400 characters.

Default:

Bitrate

Number

Bitrate for the generated music

Default: 256000

Voice Id

String

Reuse a previously uploaded voice ID

Default:

Sample Rate

Number

Sample rate for the generated music

Default: 44100

Instrumental Id

String

Reuse a previously uploaded instrumental ID

Default:
Outputs (1)

Output

Inferred

Output

Nodespell

Nodespell

London

Building the future. Join us!

nodespell.comnodespell.appNodespellAI
Creator profile

Type

Node

Status

Official

Package

Nodespell AI

Category

AI / Audio / Minimax

Input

TextAudio

Output

Audio

Keywords

Music GenerationStyle ControlLength Control
Use in Workflow