
Home

About me

Publications

Downloads

Contact

|
|
Speech Examples
Compare the proposed A/S method (aQHM) with well known speech representations.
MALE SAMPLES
SM: 30ms window length, 5ms step, 80 sinusoidal components
HNM: 2 pitch periods (p.p.) window length, 1 p.p. step, maximum voiced frequency 5500Hz
AQHM: 3 p.p. window length, 2.5ms step, maximum voiced frequency 5500Hz
STRAIGHT: Kawahara's algorithm and code
FEMALE SAMPLES
SM: 25ms window length, 5ms step, 80 sinusoidal components
HNM: 2 pitch periods (p.p.) window length, 1 p.p. step, maximum voiced frequency 5500Hz
AQHM: 3 p.p. window length, 2.5ms step, maximum voiced frequency 5500Hz
STRAIGHT: Kawahara's algorithm and code
Web site and all contents (c) Copyright Yannis Pantazis 2008, All rights reserved.
|