Improving Reverberant Speech Training Using Diffuse Acoustic Simulation

We present an efficient and realistic geometric acoustic simulation approach for generating and augmenting training data in speech-related machine learning tasks. Our physically-based acoustic simulation method is capable of modeling occlusion, specular and diffuse reflections of sound in complicate...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2020-04
Main Authors: Tang, Zhenyu, Chen, Lianwu, Wu, Bo, Yu, Dong, Manocha, Dinesh
Format: Article
Language:English
Subjects:
Acoustic simulation
Acoustics
Computer simulation
Machine learning
Occlusion
Sound
Speech recognition
Test sets
Training
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We present an efficient and realistic geometric acoustic simulation approach for generating and augmenting training data in speech-related machine learning tasks. Our physically-based acoustic simulation method is capable of modeling occlusion, specular and diffuse reflections of sound in complicated acoustic environments, whereas the classical image method can only model specular reflections in simple room settings. We show that by using our synthetic training data, the same neural networks gain significant performance improvement on real test sets in far-field speech recognition by 1.58% and keyword spotting by 21%, without fine-tuning using real impulse responses.
ISSN:2331-8422