I'm writing a course conclusion work for my college, the idea is to extract musical notes from polyphonic audio files using a artificial neural network, what would be the best method to obtain audio characteristics fast fourier transform or Mel-Frequency Cepstral Coefficients?
Vladimir Kubov
The greatest compression of musical information is achieved in *.midi format. If this does not scare you, try to understand how it is done. For example, here:
https://basicpitch.spotify.com/
https://github.com/spotify/basic-pitch/tree/main/basic_pitch
Or simplify the task to your own capabilities ;)
0 votes
0 thanks
- Badges
- Science topic
More Árlisson Alves da Silva Silveira's questions See All
Similar questions and discussions