Disentangling Speech and Non-Speech Components for Building Robust Acoustic Models from Found Data

1 code implementation • 25 Sep 2019 • Nishant Gurunath, Sai Krishna Rallabandi, Alan Black

We show that the constraints on the latent space of a VAE can be in-fact used to separate speech and music, independent of the language of the speech.

speech-recognition Speech Recognition +1