Learning Problem-agnostic Speech Representations from Multiple Self-supervised TasksEI

    Cited by: 0|Bibtex|69|

    Conference of the International Speech Communication Association, 2019.

    Abstract:

    Learning good representations without supervision is still an open issue in machine learning, and is particularly challenging for speech signals, which are often characterized by long sequences with a complex hierarchical structure. Some recent works, however, have shown that it is possible to derive useful speech representations by emplo...More
    Your rating :
    0

     

    Tags
    Comments