MLOps.community  cover image

Bringing Audio ML Models into Production // Valerio Velardo // MLOps Coffee Sessions #90

MLOps.community

00:00

Transform Spectra Into ML Spectrograms

The problem with all your data is that if you are dealing with ro oudiot, for example, a waf i, this is highly, highly dimensio dimensional. It's very difficult to take aa very long piece of ro ordio and pass it into a dep lerning modle because the dimensionality is enormous. In other words, we need to have representations of the odio that's meaninful enough, and at the same time that's compact enough so that the models can handle this information. Music processing is something so peculiar, and you have to have a set of scales that are quite unique in order to work in this space.

Play episode from 18:04
Transcript

The AI-powered Podcast Player

Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!
App store bannerPlay store banner
Get the app