电子学报 ›› 2015, Vol. 43 ›› Issue (4): 816-821.DOI: 10.3969/j.issn.0372-2112.2015.04.028

• 科研通信 • 上一篇    下一篇

基于相似关联度神经网络的音频频带扩展

刘鑫, 鲍长春   

  1. 北京工业大学电子信息与控制工程学院, 北京 100124
  • 收稿日期:2013-10-11 修回日期:2014-02-17 出版日期:2015-04-25
    • 作者简介:
    • 刘鑫 男.1986年9月出生于北京.现为北京工业大学博士研究生.主要研究方向为语音与音频信号处理.E-mail:liuxin0930@emails.bjut.edu.cn;鲍长春 男,1965年6月出生于内蒙古赤峰市,博士,教授,博士生导师,主要研究方向为语音与音频信号处理.E-mail:chchbao@bjut.edu.cn
    • 基金资助:
    • 国家自然科学基金 (No.61072089)

Audio Bandwidth Extension Method Using Similarity Correlation Degree-Based Neural Network

LIU Xin, BAO Chang-chun   

  1. School of Electronic Information and Control Engineering, Beijing University of Technology, Beijing 100124, China
  • Received:2013-10-11 Revised:2014-02-17 Online:2015-04-25 Published:2015-04-25
    • Supported by:
    • National Natural Science Foundation of China (No.61072089)

摘要:

宽带音频带宽的限制会降低其主观质量和自然度.本文提出了一种基于相似关联度神经网络的宽带向超宽带音频频带扩展方法.该方法将宽带音频的精细谱重构成多维相空间,并建立相似关联度神经网络来恢复高频成分的精细谱,同时借助高斯混合模型估计高频谱包络,并以G.722.1编码器为平台实现音频信号的带宽扩展.测试结果表明,本文方法扩展性能优于参考方法,其主观质量接近于G.722.1C超宽带编码器.

关键词: 音频编码, 音频频带扩展, 相似关联度神经网络, 相空间重构, 高斯混合模型

Abstract:

The bandwidth limitation of wideband audio degrades the subjective quality and the naturalness.In this paper,a bandwidth extension of audio signals from wideband to super-wideband was proposed by using a similarity correlation degree-based neural network.Firstly,the fine spectrum of wideband audio was converted to a multi-dimensional phase space.Then,a similarity correlation degree-based neural network was built up to reproduce the high-frequency fine spectrum.In addition,Gaussian mixture model was used to estimate the high-frequency spectral envelope.Finally,the bandwidth was extended to super-wideband by the proposed method in the ITU-T G.722.1 wideband codec.Evaluation results indicate that the proposed method is preferred over the reference methods and achieves a comparable subjective quality with the G.722.1C super-wideband codec.

Key words: audio coding, audio bandwidth extension, similarity correlation degree-based neural network, phase space reconstruction, Gaussian mixture model

中图分类号: