基于多源自适应与时序全局评分映射的音质测量
DOI:
CSTR:
作者:
作者单位:

佛山大学计算机与人工智能学院佛山528225

作者简介:

通讯作者:

中图分类号:

TN912. 35

基金项目:

广东省教育厅特色创新项目(2025KTSCX130)资助


Objective speech quality evaluation based on multi-source adaptation and temporal-global score mapping
Author:
Affiliation:

School of Computer Science and Artificial Intelligence, Foshan University, Foshan 528225,China

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    近年来,基于深度学习的无参考源语音质量客观测量(non-intrusive speech quality evaluation, NI-SQE)在语音编码、增强、识别等应用中取得了显著进展。然而,现有方法在跨域任务中的泛化能力以及捕捉语音时序特征与全局评分之间的精确映射方面仍面临挑战。为解决这些问题,提出了一种融合多源自适应与时序—全局评分映射的跨域语音质量客观测量方法。该方法通过目标域失真先验进行自适应多源选择,自适应筛选与目标域最相关的源域数据,从而减少异常源域的负迁移。在此基础上,结合卷积神经网络(convolutional neural networks, CNN)的局部感受能力和双向长短期记忆网络(bidirectional long short-term memory, BLSTM)的长时依赖建模能力,提出了基于CNN-BLSTM的音质客观评分模型,同时引入帧级预测误差与整体误差的混合损失函数,从而提高模型对帧间质量波动的鲁棒性。实验结果表明,在保持较低模型复杂度和较高推理效率的同时,在仅使用约1/2训练数据的情况下,仍优于现有主流NI-SQE方法,平均斯皮尔曼秩相关系数(spearman rank order correlation coefficient,SRCC)和皮尔逊线性相关系数(pearson linear correlation coefficient,PLCC)分别提升了2.21%~35.62%和2.61%~39.22%,说明该方法不仅具有良好的跨域泛化能力,同时具备实际工程部署的可行性。

    Abstract:

    Existing methods still face challenges in terms of generalization ability in cross-domain tasks and capturing the precise mapping between speech temporal features and global scores. To address these issues, this paper proposes a cross-domain objective speech quality evaluation method that integrates multi-source adaptation with temporal-global score mapping. The method utilizes the distortion prior of the target domain to perform adaptive multi-source selection, matching the data distributions of source and target domains, thus reducing negative transfer from anomalous source domains. On this basis, by combining the local perceptual capability of convolutional neural networks (CNN) and the long-term dependency modeling ability of bi-directional long short-term memory (BLSTM), a CNN-BLSTM-based objective speech quality scoring model is proposed. Additionally, a hybrid loss function combining frame-level prediction error with global error is introduced to improve the model’s robustness to frame-level quality fluctuations and enhance the stability of predictions. Experimental results show that the proposed method outperforms current mainstream NI-SQE approaches using only about half the training data, achieving average SRCC and PLCC improvements ranging from 2.21% to 35.62% and from 2.61% to 39.22%, respectively. This demonstrates not only its strong cross-domain generalization ability but also its feasibility for practical engineering deployment.

    参考文献
    相似文献
    引证文献
引用本文

周伟力,张悦宇,陈荟慧,杨涛,廖粤涛.基于多源自适应与时序全局评分映射的音质测量[J].电子测量与仪器学报,2026,40(9):212-221

复制
分享
相关视频

文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期: 2026-11-13
  • 出版日期:
文章二维码
×
《电子测量与仪器学报》
关于防范虚假编辑部邮件的郑重公告