Sorry, you need to enable JavaScript to visit this website.

Weighted Speech Distortion Losses for Real-time Speech Enhancement

Citation Author(s):
Ivan Tashev
Submitted by:
Yangyang Raymond Xia
Last updated:
17 May 2020 - 7:34pm
Document Type:
Presentation Slides
Document Year:
Yangyang Xia

This paper investigates several aspects of training a RNN (recurrent neural network) that impact the objective and subjective quality of enhanced speech for real-time single-channel speech enhancement. Specifically, we focus on a RNN that enhances short-time speech spectra on a single-frame-in, single-frame-out basis, a framework adopted by most classical signal processing methods. We propose two novel mean-squared-error-based learning objectives that enable separate control over the importance of speech distortion versus noise reduction. The proposed loss functions are evaluated by widely accepted objective quality and intelligibility measures and compared to other competitive online methods. In addition, we study the impact of feature normalization and varying batch sequence lengths on the objective quality of enhanced speech. Finally, we show subjective ratings for the proposed approach and a state-of-the-art real-time RNN-based method.

0 users have voted: