Virtual instruments or physical instruments? It's already used to create virtual instruments--new sounds can be created, sounds can be mixed together (e.g. Magenta).
Physically you could in theory scan/create 3d models of instruments and train it on materials, resonance, etc. and in theory it could design instruments that have greater properties you're looking for. it could generate ideas for new instruments, or improve on the capabilities of existing ones by suggesting adaptations for accessibility or personal needs of individuals. It could create hybrids of existing instruments.
But this requires you to build and train your own neural networks. As far as I'm aware, there are no existing models freely available to do this for you.
By training a neural network on a large dataset of sounds, generative AI can learn to generate new sounds that are similar to the ones in the dataset, but with unique variations and combinations. This can lead to the creation of new musical instruments that produce sounds that are unlike anything heard before.
it can be used to create new sounds and timbres that are not possible with traditional instruments.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Probability
- Probability Theory
- Reliability