Non-intrusive speech quality assessment with attention-based ResNet-BiLSTM

Speech quality is frequently affected by a variety factors in online conferencing applications, such as background noise, reverberation, packet loss and network jitter. In real scenarios, it is impossible to obtain a clean reference signal for evaluating the quality of the conferencing speech. There...

Full description

Saved in:
Bibliographic Details
Published in:Signal, image and video processing image and video processing, 2023-10, Vol.17 (7), p.3377-3385
Main Authors: Shen, Kailai, Yan, Diqun, Ye, Zhe, Xu, Xianbo, Gao, JinXing, Dong, Li, Peng, Chengbin, Yang, Kun
Format: Article
Language:English
Subjects:
Background noise
Computer Imaging
Computer Science
Image Processing and Computer Vision
Multimedia Information Systems
Original Paper
Pattern Recognition and Graphics
Quality assessment
Reference signals
Signal quality
Signal,Image and Speech Processing
Speech
Vision
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Speech quality is frequently affected by a variety factors in online conferencing applications, such as background noise, reverberation, packet loss and network jitter. In real scenarios, it is impossible to obtain a clean reference signal for evaluating the quality of the conferencing speech. Therefore, an effective non-intrusive speech quality assessment (NISQA) method is necessary. In this paper, we propose a new network framework for NISQA based on ResNet and BiLSTM. ResNet is utilized to extract local features, while BiLSTM is used to integrate representative features with long-term time dependencies and sequential characteristics. Considering that ResNet may result in the loss of context information when applied to the NISQA task, we propose a variant of ResNet which can preserve the time series information of the conferencing speech. The experimental results demonstrate that the proposed method has a high correlation with the mean opinion score of clean, noisy and processed speech.
ISSN:1863-1703
1863-1711
DOI:10.1007/s11760-023-02559-2