Skip to main navigation Skip to search Skip to main content

Pitch transformation in neural network based voice conversion

  • Feng Long Xie
  • , Yao Qian
  • , Frank K. Soong
  • , Haifeng Li
  • Harbin Institute of Technology
  • Microsoft USA

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

Abstract

In voice conversion task, prosody conversion especially pitch conversion is a very challenging research topic because of the discontinuity property of pitch. Conventionally pitch conversion is always achieved by adjusting the mean and variance of the source pitch distribution to the target pitch distribution. This method removes most of the detailed information of the speaker's prosody and only maintains the global F0 contour. In this paper, we propose a neural network based pitch conversion system which converts F0 and spectral features all together frame by frame. Experimental results show that neural network based pitch conversion can significantly reduce the Unvoiced/Voiced error and RMSE of F0 between converted pitch and target pitch compared with the conventional Gaussian normalized transformation method. Wavelet decomposition for F0 can further improve the performance of voice conversion.

Original languageEnglish
Title of host publicationProceedings of the 9th International Symposium on Chinese Spoken Language Processing, ISCSLP 2014
EditorsMinghui Dong, Jianhua Tao, Haizhou Li, Thomas Fang Zheng, Yanfeng Lu
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages197-200
Number of pages4
ISBN (Electronic)9781479942206
DOIs
StatePublished - 24 Oct 2014
Event9th International Symposium on Chinese Spoken Language Processing, ISCSLP 2014 - Singapore, Singapore
Duration: 12 Sep 201414 Sep 2014

Publication series

NameProceedings of the 9th International Symposium on Chinese Spoken Language Processing, ISCSLP 2014

Conference

Conference9th International Symposium on Chinese Spoken Language Processing, ISCSLP 2014
Country/TerritorySingapore
CitySingapore
Period12/09/1414/09/14

Keywords

  • neural network
  • pitch
  • voice conversion

Fingerprint

Dive into the research topics of 'Pitch transformation in neural network based voice conversion'. Together they form a unique fingerprint.

Cite this