A solo maker wants to train a small AI model for voice-made music

A music-making tool is being built for people who do not know how to play an . A person can hum a melody and get it back as real sounds, or describe a bassline and have one created to match the song’s tempo and key. The goal is similar to coding , but for music: use voice and plain instructions instead of manual music skills.

The main open question is what it would take to train a small AI model for this exact job. The hard parts are choosing the right music data, picking a training approach, finding useful reading or experts, and learning from people who have already worked on music generation or .

Key points

  • The tool aims to turn humming or spoken music ideas into usable parts.
  • It should create melodies and basslines that fit the song’s tempo and key.
  • The maker wants a small AI model focused on this one music task.
  • The biggest unknowns are data, training method, and audio model know-how.
  • For a solo maker, using existing models first may be more practical than training from scratch.
Read original