AIMLAPI Music Generation

Overview

Generate music tracks using state-of-the-art AI models (Suno, Udio, Minimax, ElevenLabs).

Quick start

# General music (instrumental)
python {baseDir}/scripts/gen_music.py \
  --prompt "cyberpunk synthwave with heavy bass and retro synths" \
  --model "minimax/music-2.0"

# Song with lyrics
python {baseDir}/scripts/gen_music.py \
  --prompt "A happy pop song about a robot learning to feel" \
  --lyrics "[Verse 1]\nWires and gears, clicking in time..." \
  --model "minimax/music-2.0"

# Short clip (ElevenLabs)
python {baseDir}/scripts/gen_music.py \
  --prompt "lo-fi pop hip-hop ambient" \
  --model "elevenlabs/eleven_music" \
  --length 20000

Arguments

• --prompt: (Required) Style or context for the music.
• --lyrics: Optional lyrics for vocal tracks.
• --model: Model choice (default: minimax/music-2.0).
• --length: Length in milliseconds (primarily for ElevenLabs).
• --out-dir: Directory to save the final MP3.

Workflow

The script uses a two-step process:

1. POST /v2/generate/audio: Creates the generation task.
2. GET /v2/generate/audio?generation_id=...: Polls for the result until completed or failed.