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

Towards an ASR Approach Using Acoustic and Language Models for Speech Enhancement

Citation Author(s):
Submitted by:
KHANDOKAR MD NAYEM
Last updated:
24 June 2021 - 4:06pm
Document Type:
Presentation Slides
Document Year:
2021
Event:
Presenters:
Khandokar Md. Nayem
Paper Code:
SPE-51.6
 

Recent work has shown that deep-learning based speech enhancement performs best when a time-frequency mask is estimated. Unlike speech, these masks have a small range of values that better facilitate regression-based learning. The question remains whether neural-network based speech estimation should be treated as a regression problem. In this work, we propose to modify the speech estimation process, by treating speech enhancement as a classification problem in an ASR-style manner. More specifically, we propose a quantized speech prediction model that classifies speech spectra into a corresponding quantized class. We then train and apply a language-style model that learns the transition probabilities of the quantized classes to ensure more realistic speech spectra. We compare our approach against time-frequency masking approaches, and the results show that our quantized spectra approach leads to improvements.

up
0 users have voted: