SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition

We present SpecAugment, a simple data augmentation method for speech recognition. SpecAugment is applied directly to the feature inputs of a neural network (i.e., filter bank coefficients). The augmentation policy consists of warping the features, masking blocks of frequency channels, and masking bl...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	arXiv.org 2019-12
Hauptverfasser:	Park, Daniel S, Chan, William, Zhang, Yu, Chung-Cheng, Chiu, Barret Zoph, Cubuk, Ekin D, Le, Quoc V
Format:	Artikel
Sprache:	eng
Schlagworte:	Automatic speech recognition Computer Science - Computation and Language Computer Science - Learning Computer Science - Sound Data augmentation Filter banks Hybrid systems Masking Neural networks Statistics - Machine Learning Switching theory Voice recognition
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!