Three-class Overlapped Speech Detection using a Convolutional Recurrent Neural Network

• Published in: arXiv.org
• Main Authors: Jee-weon Jung, Hee-Soo Heo, Kwon, Youngki, Joon Son Chung, Bong-Jin, Lee
• Format: Paper
• Language: English
• Published: Ithaca Cornell University Library, arXiv.org 07.04.2021
• Subjects: Computer architecture; Neural networks; Recall; Recurrent neural networks; Speech

In this work, we propose an overlapped speech detection system trained as a three-class classifier. Unlike conventional systems that perform binary classification as to whether or not a frame contains overlapped speech, the proposed approach classifies into three classes: non-speech, single speaker...