Noise Reduction Method Focusing on Spectral Envelopment and Fine Structure of Speech

• Fukawa Taichi

  Graduate School of Engineering, Tokyo City University

• Jin'no Kenya

  Graduate School of Engineering, Tokyo City University

抄録

We propose a method to enhance noise reduction performance by separating a speech spectrum into spectral envelopes and fine structures using cepstrum analysis and linear predictive coding (LPC) analysis, and removing noise using an autoencoder (AE). A technique for removing noise from the spectrum of noise-containing speech is to use AE to reconstruct the spectrum of speech through the latent variables of the speech. We focused on spectral envelopes and fine structures that constitute speech, and improved the independence between latent variables in AE to reconstruct the speech spectrum by separating them in advance. In this way, we confirmed that the performance of noise reduction was improved in exchange for a slight decrease in the reproducibility of speech spectra when cepstrum analysis was used. It was also confirmed that cepstrum analysis was superior to LPC analysis in noise reduction.

収録刊行物

• 信号処理

  信号処理 25 (6), 233-237, 2021-11-01

  信号処理学会

キーワード

• noise reduction
• spectral envelope
• fine structure
• autoencoder
• cepstrum analysis
• LPC analysis

詳細情報

• CRID: 1390008445626226304
• NII論文ID: 130008110099
• DOI: 10.2299/jsp.25.233
• ISSN: 18801013
• ISSN: 13426230
• Web Site: https://www.jstage.jst.go.jp/article/jsp/25/6/25_233/_pdf
• 本文言語コード: en
• データソース種別:
  • JaLC
  • Crossref
  • CiNii Articles
  • KAKEN
• 抄録ライセンスフラグ: 使用不可