Presentation 2009-01-30
Clustering to make a set of templates for voicing rate and pitch determination
Yosuke KATAGUCHI, Hideo MIYABAYASHI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In our past voiced/unvoiced detection, we have classified speech sections with two values. In this study, we introduce voicing rate of multiple values to express mixture ratio of a voiced/unvoiced speech signal. We propose a simple, easy and effective method that determinates voicing rate through two-stage clustering with Euclid distance and angular distance between speech features vectors extracted from real speech by bank of BandPass Filter Pairs (BPFP). Then we could make a set of templates for voicing rate determination. In addition, it is shown that we examine a speech pitch determination method and succeed in determination precision be not inferior to methods using artificial standard vectors.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) voicing rate / pitch / BPFP / clustering
Paper # SP2008-135
Date of Issue

Conference Information
Committee SP
Conference Date 2009/1/22(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Speech (SP)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Clustering to make a set of templates for voicing rate and pitch determination
Sub Title (in English)
Keyword(1) voicing rate
Keyword(2) pitch
Keyword(3) BPFP
Keyword(4) clustering
1st Author's Name Yosuke KATAGUCHI
1st Author's Affiliation Toyama National College of Maritime Technology()
2nd Author's Name Hideo MIYABAYASHI
2nd Author's Affiliation Toyama National College of Maritime Technology
Date 2009-01-30
Paper # SP2008-135
Volume (vol) vol.108
Number (no) 422
Page pp.pp.-
#Pages 6
Date of Issue