한국음향학회지 (The Journal of the Acoustical Society of Korea)

  • 제21권8호
  • /
  • Pages.727-734
  • /
  • 2002
  • /
  • 1225-4428(pISSN)
  • /
  • 2287-3775(eISSN)
한국음향학회 (The Acoustical Society of Korea)
권호 표지

유성음 구간 검출을 위한 간단한 알고리즘에 관한 연구

A Study on the Simple Algorithm for Discrimination of Voiced Sounds

  • 장규철 (한국과학기술원 전자전산학과) ;
  • 우수영 (한국과학기술원 전자전산학과) ;
  • 박용규 (한국과학기술원 전자전산학과) ;
  • 유창동 (한국과학기술원 전자전산학과)
  • 발행 : 2002.11.01
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

본 논문에서는 유ㆍ무성음 구간을 검출하기 위한 간단한 알고리즘을 제안한다. 제안된 방법은 음성의 유ㆍ무성음의 주기성에 대한 특성을 보완할 수 있는 저대역 에너지와 영교차율, 그리고 주기성의 안정성을 판단하기 위한 피치 변화량을 파라미터로 사용하였다. 유ㆍ무성음의 구간검출을 음소단위의 검출이라는 측면에서 접근하여 음소군의 검출율과 음소군내의 음소의 검출율을 얻었다. TIMIT코퍼스 (corpus)를 데이터베이스로 사용하여 실험했을 때 유성음 음소 검출율이 약 13% 향상되었다.

A simple algorithm for discriminating voiced sounds in a speech is proposed in this paper. In addition to low-frequency energy and zero-crossing rate (ZCR), both of which have been widely used in the past for identifying voiced sounds, the proposed algorithm incorporates pitch variation to improve the discrimination rate. Based on TIMIT corpus, evaluation result shows an improvement of 13% in the discrimination of voiced phonemes over that of the traditional algorithm using only energy and ZCR.

키워드

참고문헌

  1. Digital Processing of Speech Signals L.Rabiner;R.W.Schafer
  2. TIMIT, Acoustic-Phonetic Continuous Speech Corpus CD, American Helix
  3. IEEE Trans. Acoust. Speech. Signal Processing v.ASSP-24 A pattern recognition approach to voiced-unvoiced-silence classification with applications to speech recognition B.S.Atal;L.R.Rabiner
  4. IEEE Trans. Acoust. Speech Signal Processing v.ASSP-25 A pitch extraction algorithm based on LPC inverse filtering and AMDF C.K.Un;S.C.Yang
  5. ICASSP-90 Neural Networks for Voiced/Unvoiced Speech Classification A.Bendiksen;K.Steiglitz
  6. ICASSP-91 Robust Voiced/Unvoiced Speech Claasification using a Neural Net R.P.Cohn
  7. ICASSP-91 Fast Neural Net Training Algorithm and Its Application to Voiced-Unvoiced-Silence Classification of Speech T.G.Crippa;A.E.Jaroudi
  8. ISSPA-96 v.1 Voiced/Unvoiced/Silence Classification of Speech Using 2-Stage Neural Networks With Delayed Decision Input R.Ahn;W.H.Holmes
  9. TIA/EIA/IS-127 Enhanced variable rate codec, speech service option 3 for wideband spread spectrum digital systems
  10. IEEE Trans. Acoust. Speech, Signal Processing v.ASSP-28 A procedure for using pattern classification techniques to obtain a voiced/unvoiced classifier L.J.Siegel
  11. IEEE Trans. Acoust. Signal Processing v.39 An Autocorrelation Pitch Detector and Voicing Decision with Cinfidence Measures Developed for Noise-Corrupted Speech D.A.Krubsack;R.J.Njederjohn https://doi.org/10.1109/78.80814