Speech bandwidth extension based on GMM and clustering method

Yingxue Wang, Shenghui Zhao, Yingying Yu, Jingming Kuang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

14 Citations (Scopus)

Abstract

Conventional Gaussian mixture model (GMM) Speech Bandwidth Extension (BWE) methods often suffer from the overly smoothed problem. Thus, a method of BWE based on a cluster process and GMM whose parameters are determined by expectation-Maximization (EM) is proposed. Firstly, a cluster process is used to cluster the low frequency and high frequency parameters, and then the GMM for each cluster is established. Later on, the parameters of low frequency are transformed to the parameters of high frequency according to the learned mapping function of the corresponding GMM. Self-organization Feature Mapping (SOFM) and Vector Quantization (VQ) are applied as the cluster. It is shown by subjective evaluation and objective evaluation that, the proposed method improves the quality of the synthesized speech signals compared with the conventional GMM-based BWE method and overcomes the over-smoothed problem caused by the traditional GMM-based BWE method largely.

Original languageEnglish
Title of host publicationProceedings - 2015 5th International Conference on Communication Systems and Network Technologies, CSNT 2015
EditorsGeetam Singh Tomar
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages437-441
Number of pages5
ISBN (Electronic)9781479917976
DOIs
Publication statusPublished - 28 Sept 2015
Event5th International Conference on Communication Systems and Network Technologies, CSNT 2015 - Gwalior, India
Duration: 4 Apr 20156 Apr 2015

Publication series

NameProceedings - 2015 5th International Conference on Communication Systems and Network Technologies, CSNT 2015

Conference

Conference5th International Conference on Communication Systems and Network Technologies, CSNT 2015
Country/TerritoryIndia
CityGwalior
Period4/04/156/04/15

Keywords

  • Bandwidth extension
  • Gaussian mixture model
  • Self-organizing feature m Vector Quantization

Fingerprint

Dive into the research topics of 'Speech bandwidth extension based on GMM and clustering method'. Together they form a unique fingerprint.

Cite this